Scatterplots and Line of Best Fit - Complete Interactive Lesson
Part 1: Data Analysis
Problem Solving: Ratios, Rates & Proportions
Part 1 of 7 — Setting Up and Solving Proportions
This is one of the most heavily tested topics on the SAT Math section. About 25-30% of Math questions fall under Problem Solving & Data Analysis.
Ratios
A ratio compares two quantities: If a recipe uses 3 cups flour to 2 cups sugar, the ratio is 3:2 or 3/2.
Setting Up Proportions
Cross-multiply to solve:
53=20ximplies3×20=
Unit Rates
A unit rate has a denominator of 1:
240 miles in 4 hours → 60 mph
$45 for 3 shirts → $15 per shirt
SAT Trap: Mixing Up Parts and Wholes
If the ratio of boys to girls is 3:5, there are 8 total parts (not 5).
Boys = 3/8 of total
Girls = 5/8 of total
Dimensional Analysis
Convert units by multiplying fractions:
60hourmiles×60 min
Deep Dive: Complex Ratio & Proportion Problems
Worked Example 1: Multi-Step Ratio
Step
Work
Problem
"In a mixture, the ratio of water to concentrate is 5:2. If there are 21 total cups, how much water is needed?"
Total parts
5+2=7 parts
Each part
21÷7 cups per part
Advanced Ratio & Proportion Problems 🎯
Ratio & Proportion Setup — Select the correct approach.
Part 1 Summary: Ratios, Rates & Proportions
Concept
Formula
SAT Trap
Ratio a:b
Part =a+b total
Part 2: Scatterplots
Percentages: Increase, Decrease & Applications
Part 2 of 7 — Mastering Percent Problems
Percent Formula
Percent=WholePart×100
Percent Increase/Decrease
Part 3: Probability
Two-Way Tables & Data Interpretation
Part 3 of 7 — Reading Tables and Finding Probabilities
Two-Way Tables
These organize data by two categories. Example:
Freshman
Sophomore
Total
Male
120
100
220
Female
130
150
280
Total
250
250
500
Conditional Probability from Tables
"What fraction of sophomores are female?"
Look at the Sophomore column: 150 female out of 250 total = 150/250 = 3/5
"Given that" = Restrict to a Subgroup
"Given that a student is male, what is the probability they are a freshman?"
Restrict to Male row: 120 freshman out of 220 male = 120/220 = 6/11
Marginal vs. Conditional
Marginal: P(Female) = 280/500 — uses the grand total
Conditional: P(Female | Sophomore) = 150/250 — uses a column/row total
Association vs. Independence
Part 4: Two-Way Tables
Statistics: Center, Spread & Shape
Part 4 of 7 — Mean, Median, Standard Deviation
Measures of Center
Mean = sum of all values / count. Sensitive to outliers.
Median = middle value when sorted. Resistant to outliers.
When to Use Mean vs. Median
Symmetric data → mean ≈ median, use either
Skewed data or outliers → median is more representative
Standard Deviation
Measures how spread out data is from the mean.
Low SD → data points close to mean (consistent)
High SD → data points far from mean (variable)
You won't calculate SD on the SAT, but you must compare SDs:
{10, 10, 10, 10, 10} → SD = 0 (no spread)
{8, 9, 10, 11, 12} → small SD
{1, 3, 10, 17, 19} → large SD
Effect of Adding/Removing Values
Adding a value equal to the mean → mean unchanged, SD decreases
Adding an outlier → mean shifts toward outlier, SD increases
Removing an outlier → mean moves away from outlier, SD decreases
Shape of Distributions
Right-skewed (tail to right): mean > median
Left-skewed (tail to left): mean < median
Symmetric: mean ≈ median
Deep Dive: Statistics in Action
Part 5: Statistical Modeling
Scatterplots & Line of Best Fit
Part 5 of 7 — Interpreting Trends and Making Predictions
Reading Scatterplots
Positive association: as x increases, y increases (upward trend)
Negative association: as x increases, y decreases (downward trend)
No association: no visible pattern
Line/Curve of Best Fit
The line that minimizes the total distance from all points. Key interpretations:
Slope = rate of change (For each 1-unit increase in x, y changes by [slope])
y-intercept = predicted y-value when x = 0
Making Predictions
Use the equation to predict values:
If y = 2.3x + 15 models study hours vs. test score:
10 hours → predicted score: 2.3(10) + 15 = 38
Interpolation vs. Extrapolation
Interpolation (within data range): reliable predictions
Extrapolation (beyond data range): unreliable — the trend may not continue
Residuals
Residual = actual – predicted
Positive residual: actual is above the line
Negative residual: actual is below the line
Random residuals → good model
Patterned residuals (curved) → wrong model type
Deep Dive: Scatterplot Analysis
Worked Example 1: Interpreting Slope in Context
Step
Part 6: Problem-Solving Workshop
Probability & Counting
Part 6 of 7 — SAT Probability Essentials
Basic Probability
P(event)=total outcomesfavorable outcomes
Complement Rule
Part 7: Review & Applications
Problem Solving & Data Review
Part 7 of 7 — Mixed Practice & Strategy
Topic Checklist
✓ Ratios, rates, proportions, and unit conversion
✓ Percent increase/decrease and successive changes
✓ Two-way tables and conditional probability
✓ Mean, median, standard deviation, and outliers
✓ Scatterplots, line of best fit, and residuals
✓ Probability, expected value, and counting
SAT Strategy for This Section
Read the question last — scan the table/graph first to understand the data
Identify what the denominators should be — marginal vs. conditional probability
Watch for traps: part-to-part vs. part-to-whole ratios
Use estimation — if a scatterplot has a clear trend, estimate before calculating
Common Mistakes
Confusing "percent increase" with "percentage points"
Using the wrong total for conditional probability
Forgetting that percent change compounds (not additive)
Extrapolating beyond the data range when the question asks for interpolation
Deep Dive: Mixed SAT Data Problems
Worked Example 1: Multi-Concept Problem
Step
Work
Problem
"A dataset's mean is 50 and SD is 8. Every value is doubled then 10 is added. Find the new mean and SD."
Double
Mean , SD
5ximpliesx=
12
1 hour
=
1minmile
=
3
Water
5×3=15 cups
Worked Example 2: Unit Conversion Chain
Step
Work
Problem
"A printer prints 12 pages per minute. How many pages in 2.5 hours?"
Convert hours → minutes
2.5×60=150 minutes
Calculate
12×150=1,800 pages
Ratio vs. Fraction — Key Difference
Statement
Ratio
Fraction of Total
"Boys to girls is 3:5"
3:5
Boys =83, Girls =85
"Boys to total is 3:8"
3:8
Boys =83
"3 out of every 5 are boys"
3:2 (boys:girls)
Boys =53
Dimensional Analysis — Multi-Step
Convert 45 mph to feet per second:
45hrmi×1 mi5280 ft×3600 sec1 hr=66secft
a
×
Confusing part:part with part:whole
Cross multiplication
ba=dc → ad=bc
Setting up the wrong proportion
Unit rate
1 unitquantity
Not reducing to denominator of 1
Dimensional analysis
Cancel matching units
Missing a conversion step
Next: Percentages — increase, decrease, and successive changes →
% Change=OriginalNew−Original×100
Shortcut multipliers:
20% increase → multiply by 1.20
15% decrease → multiply by 0.85
8% tax → multiply by 1.08
Successive Percent Changes
A 10% increase followed by a 10% decrease is NOT back to the original:
100+10%110−10%99
SAT Classic: "What percent of X is Y?"
Translate directly: "What percent of 80 is 24?"
8024×100=30%
Percent vs. Percentage Points
"Increased from 40% to 52%" = increase of 12 percentage points but a 30% increase (12/40 × 100).
Deep Dive: Percent Problem Strategies
Worked Example 1: Finding Original Price
Step
Work
Problem
"After a 30% discount, a jacket costs $56. What was the original price?"
Setup
You pay 70% of original: 0.70x=56
Solve
x=56/0.70=80
Common mistake
Adding 30% of 56: 56+16.80=72.80 ← WRONG
Worked Example 2: Successive Changes
Step
Work
Problem
"A stock rises 25% one year, then drops 20% the next. Net change?"
Year 1
100×1.25=125
Year 2
125×0.80=100
Percent Multiplier Quick Reference
Phrase
Multiplier
Example
15% increase
×1.15
200×1.15=230
15% decrease
×0.85
"Percent OF" vs. "Percent MORE THAN"
"A is 25% of B" → A=0.25B
"A is 25% more than B" → A=1.25B
"A is 25% less than B" → A=0.75B
Advanced Percent Problems 🎯
Percent Multiplier Check — Select the correct multiplier.
Part 2 Summary: Percentages
Concept
Formula
Percent of
WholePart×100
Percent change
OldNew−Old×100
x% increase
Multiply by (1+x/100)
x% decrease
Multiply by (1−x/100)
Successive changes
Multiply the multipliers
Finding original
Divide by the multiplier
SAT Traps
Successive equal percent changes DON'T cancel out
"A is 60% more than B" ≠ "B is 60% less than A"
Always divide by the original for percent change
Next: Two-way tables and data interpretation →
Two variables are independent if knowing one doesn't change the probability of the other.
If P(Female) = P(Female | Sophomore), gender and class year are independent
If those probabilities differ, there's an association
Deep Dive: Navigating Two-Way Tables
Worked Example 1: Filling In a Table
Step
Work
Problem
"200 employees: 120 full-time, 80 part-time. 90 have benefits; of those, 75 are full-time. Complete the table."
Full-time + benefits
75
Full-time, no benefits
120−75=45
Part-time + benefits
90−75=15
Part-time, no benefits
80−15=65
Benefits
No Benefits
Total
Full-time
75
45
120
Part-time
15
65
80
Total
90
110
200
Worked Example 2: Testing for Independence
Step
Work
Question
"Is having benefits independent of employment type?"
P(Benefits)
90/200=0.45
**P(Benefits
Full-time)**
Compare
0.45=0.625 → NOT independent
Denominator Guide
Question Phrasing
Denominator
"What fraction of ALL students...?"
Grand total
"What fraction of males...?"
Row total (Males)
"What fraction of freshmen...?"
Column total (Freshman)
"Among those who passed..."
Subtotal of those who passed
SAT Trap: Joint vs. Conditional
Joint: P(male AND freshman) =120/500 (out of everyone)
Conditional: P(freshman | male) =120/220 (out of males only)
Advanced Two-Way Table Problems 🎯
Pick the Right Denominator — What goes in the denominator for each question?
Part 3 Summary: Two-Way Tables
Concept
Key Fact
Marginal probability
Uses the grand total as denominator
Conditional probability
Restricts to a row or column total
Joint probability
One specific cell ÷ grand total
Independence test
P(A) = P(A
Filling in tables
Rows and columns must sum to their totals
SAT Strategy
Read the question word-for-word to find the correct denominator.
"Given that" or "among" = conditional → use a subtotal.
"Of all" = marginal → use the grand total.
Next: Statistics — mean, median, and standard deviation →
Worked Example 1: Finding a Missing Value
Step
Work
Problem
"Five test scores have mean 82. The first four are 78, 85, 92, 71. What is the fifth score?"
Total needed
82×5=410
Sum of four
78+85+92+71=326
Fifth score
410−326=84
Worked Example 2: Effect of Removing a Value
Step
Work
Problem
"Data: {10, 12, 14, 15, 100}. How do mean and median change if 100 is removed?"
With 100
Mean =151/5=30.2, Median =14
Without 100
Mean =51/4=12.75, Median =13
Effect
Mean drops significantly (30.2→12.75), median barely changes (14→13)
Adding a Constant vs. Multiplying
Operation
Effect on Mean
Effect on Median
Effect on SD
Add k to all values
Mean +k
Median +k
SD unchanged
Multiply all by k
Mean ×k
Median ×k
SD $\times
SAT favorite: "If every student's score increases by 5 points, what happens to the standard deviation?" → Nothing — adding a constant shifts all values equally.
Skewness Quick Reference
Shape
Tail Direction
Relationship
Example
Right-skewed
Long tail right
Mean > median
Income distribution
Left-skewed
Long tail left
Mean < median
Easy test scores
Symmetric
Equal tails
Mean ≈ median
Heights in a population
Advanced Statistics Problems 🎯
Statistics Quick Check — Select the correct answer.
Part 4 Summary: Statistics
Measure
What It Tells You
Sensitive to Outliers?
Mean
Average value
YES
Median
Middle value
NO
SD
Spread from mean
YES
Range
Max − Min
YES
Key Rules
Add constant k: mean & median shift by k, SD unchanged
Multiply by k: mean, median, & SD all multiply by ∣k∣
Right-skewed → mean > median
Outlier → use median as the better center
Next: Scatterplots and line of best fit →
Work
Model
y=3.5x+120 where x = years of experience, y = weekly earnings ($)
Slope meaning
For each additional year of experience, weekly earnings increase by $3.50.
y-intercept
A worker with 0 years of experience earns $120/week.
SAT phrasing
"The estimated increase in weekly earnings for each additional year of experience"
Worked Example 2: Choosing the Best Model
Data Pattern
Best Model
How to Tell
Straight upward trend
Linear (y=mx+b)
Residuals are random
Curve (increasing rate)
Exponential (y=abx)
Residuals show U-pattern for linear
Curve (decreasing rate)
Logarithmic or square root
Curve levels off
Ups and downs
Quadratic (y=ax2+bx+c)
Parabolic residual pattern
Correlation Coefficient (r)
r Value
Strength
Direction
r=1.0
Perfect
Positive
0.7<r<1.0
Strong
Positive
0.3<r<0.7
Moderate
Positive
0<r<0.3
Weak
Positive
r=0
None
—
r<0
Same scale, opposite direction
Negative
SAT key fact:r2 = proportion of variation in y explained by x. If r=0.8, then r2=0.64, meaning 64% of the variation is explained.
Advanced Scatterplot Problems 🎯
Scatterplot Interpretation — Select the correct answer.
Part 5 Summary: Scatterplots & Best Fit
Concept
Key Fact
Slope
Rate of change in context
y-intercept
Predicted value when x=0
Residual
Actual − predicted
r
Strength and direction of linear relationship
r2
Proportion of variation explained
Random residuals
Good model fit
Patterned residuals
Try different model type
Interpolation
Reliable (within data range)
Extrapolation
Unreliable (beyond data range)
Next: Probability and expected value →
P(not A)=1−P(A)
Often easier: "What's the probability of getting AT LEAST one?" = 1 − P(none).
"Or" vs. "And"
P(A or B) = P(A) + P(B) − P(A and B)
P(A and B) for independent events = P(A) × P(B)
SAT Probability Questions — Common Types
From a table: "A randomly selected student from the table above..."
Cards/marbles: "If 3 red and 5 blue marbles..."
Surveys: "Based on the survey results, what proportion..."
Expected Value
If the SAT asks what value you'd "expect":
Expected=Total×P(event)
Example: 200 people surveyed, 35% prefer A → Expected = 200 × 0.35 = 70
Relative Frequency
Just another word for proportion:
Relative frequency of A=total countcount of A
Deep Dive: Probability Strategies
Worked Example 1: "At Least One" with Complement
Step
Work
Problem
"Roll a die twice. What is P(at least one 6)?"
Complement
P(no sixes) =65×65=3625
Answer
P(at least one 6) =1−3625=36
Worked Example 2: Expected Value from a Survey
Step
Work
Problem
"In a sample, 42% prefer Brand A. If 500 people are surveyed from the same population, how many would you expect to prefer Brand A?"
Expected
500×0.42=210 people
Probability Rules Summary
Rule
Formula
When to Use
Complement
P(not A)=1−P(A)
"At least one", "none"
Or (general)
SAT Probability from Tables
The SAT usually gives you a two-way table and asks:
Simple probability: one cell ÷ grand total
Conditional: one cell ÷ row/column total
"At least": use complement (1 − P(none))
Advanced Probability Problems 🎯
Probability Rule Identification — Match the scenario to the correct approach.
Part 6 Summary: Probability
Rule
Formula
Key Word
Basic
Favorable ÷ Total
"probability of"
Complement
1−P(A)
"at least one", "not"
Or (exclusive)
P(A)+P(B)
Can't happen together
Or (overlap)
P(A)+P(B)−P(A∩B)
Can happen together
And (independent)
P(A)×P(B)
With replacement, separate trials
And (dependent)
$P(A) \times P(B
A)$
Expected value
Total × P(event)
"How many would you expect"
Next: Comprehensive review and mixed practice →
=50×2=100
=8×2=16
Add 10
Mean =100+10=110, SD =16 (unchanged by adding)
Answer
New mean =110, new SD =16
Worked Example 2: Comprehensive Table + Probability
Step
Work
Problem
"150 students surveyed: 60 prefer A, 50 prefer B, 40 prefer C. Of the A-preferrers, 40 are juniors. What is P(Junior
Restrict
Given "prefers A" → denominator =60
Answer
6040=32
SAT Problem Solving Cheat Sheet
Topic
Key Formula
Common Trap
Ratios
Part =a+ba× total
Part:part vs. part:whole
Percents
Multiplier method
Successive changes compound
Two-way tables
Conditional → use subtotal
Wrong denominator
Mean
nsum
Outliers distort
SD
Spread from mean
Add constant → SD unchanged
Scatterplots
Slope = rate of change
Extrapolation ≠ interpolation
Probability
Complement for "at least one"
With vs. without replacement
Time Management for This Section
Difficulty
Time Budget
Strategy
Easy (direct read from table)
30 sec
Read carefully, answer
Medium (one calculation)
60 sec
Set up, solve, check
Hard (multi-step)
90 sec
Plan approach first
Very hard (trap question)
90+ sec
Skip, flag, return
SAT Problem Solving Challenge 🎯
Problem Solving Quick Check — Select the correct answer.
Full Topic Summary: Problem Solving & Data
Part
Topic
Must-Know
1
Ratios & Proportions
Part:whole, cross-multiply, unit rates
2
Percentages
Multiplier method, successive changes compound
3
Two-Way Tables
Marginal vs. conditional vs. joint probability
4
Statistics
Mean/median/SD, outlier effects, skewness
5
Scatterplots
Slope in context, residuals, r and r2
6
Probability
Complement, OR/AND rules, expected value
7
Review
Decision framework, time management, traps
Top Strategies
Read the question carefully — identify what the denominator should be
Use multipliers for percent problems
Complement for "at least one" probability
Median when data has outliers
Check your answer — does it make sense in context?
🎉 Problem Solving & Data complete!
Net change
0% — it returned to the original!
Shortcut
1.25×0.80=1.00 — multiply the multipliers
200×0.85=170
6% tax on top
×1.06
50×1.06=53
40% of
×0.40
0.40×80=32
Triple (200% increase)
×3.00
10×3=30
Conclusion
Full-time employees are more likely to have benefits → there IS an association.