SAT Data Analysis

Topic: Statistics, Rates, Percentages, and Probability

Joey Nguyen

1 Introduction: The Power of Data in the Real World

Every time Netflix recommends a movie, algorithmic platforms curate your feed, or epi-
demiologists evaluate a vaccine, they are using Data Analysis and Statistics. We live
in an information-rich era; the ability to interpret numbers, parse trends, and separate
signal from noise is an invaluable cognitive tool.
On the Digital SAT, the Problem Solving and Data Analysis module accounts
for nearly 30% of your final score. Furthermore, in the modern Vietnamese National
Curriculum, statistics and combinatorial probability have been significantly expanded to
bridge the gap between classic high school algebra and real-world data science. This
document covers every fundamental rule up to advanced dispersion models.

2 Units, Rates, and Ratios

Before extracting global inferences from large datasets, an analyst must first control
individual quantities and measurement conversions with perfect precision.

2.1 1. Dimensional Analysis (Unit Conversion)

The SAT frequently demands rapid transitions between distinct compound measurement
scales (e.g., converting volumetric rates or compound speeds).

The Identity Fraction Method (Chain Rule)

To secure clean unit conversions without conceptual errors, multiply the target
value by formal identity fractions (fractions equal to 1), structurally ensuring
unwanted operational units cancel out diagonally. Core Rule: Target Value ×
Desired Unit
Old Unit
.

Example Problem: A machine discharges fuel at an operating speed of 60 miles

per hour (mph). Convert this delivery vector into feet per second (ft/s), assuming that
1 mile = 5280 feet.
60 miles 5280 feet 1 hour 1 minute 60 × 5280
× × × = = 88 ft/s (1)
1 hour 1 mile 60 minutes 60 seconds 3600

2.2 2. Rates and Proportions

A rate is a directional ratio comparing two unique quantities measured in non-identical
physical dimensions (e.g., pressure per square unit, currency valuation over time).
 • Average Speed Formula: Average Speed = Total Cumulative Distance
Total Operational Time
.

• Critical SAT Trap Alert! The average velocity is NEVER the simple arithmetic
mean of the individual speed values. If an object moves 40 mph to an endpoint and
returns along the exact same path at 60 mph, its total harmonic average speed is
48 mph, not 50 mph!

3 Percentages and Percent Change

Percentages represent fractions normalized out of a base value of 100. The SAT reg-
ularly conceals simple relationships within multi-tiered commercial or financial system
narratives.

3.1 1. Linear Percentage Translation Rules

To map word questions into standard algebraic properties:

• ”What percent P of X is equal to Y ?” =⇒ P

100
· X = Y =⇒ P = Y
X
· 100

• ”A base quantity X is augmented by 20%” =⇒ X · (1 + 0.20) = 1.20X

• ”A base quantity X is discounted by 15%” =⇒ X · (1 − 0.15) = 0.85X

3.2 2. Percent Change Dynamics

New Measured Value − Original Base Value
Percent Change = × 100% (2)
Original Base Value

3.3 3. Interest Metrics: Simple vs. Compound

• Simple Interest Balance: A = P (1 + rt)
Interest yield is computed exclusively on the pristine initial principal capital P .
 nt
• Compound Interest Balance: A = P 1 + nr
Interest accrues directly on the combined total of the initial principal plus all pre-
viously accumulated interest, updated n times over total period t.

4 Descriptive Statistics: Measures of Central Ten-

dency
Central tendency metrics summarize a complex matrix of numerical entries using a single
representative value.

• Mean
P (x̄): The balanced arithmetic average of the collection. Let n be total items:
n
xi
x̄ = i=1
n
.

• Median: The precise spatial midpoint value when a distribution is ordered linearly
from least to greatest. If the number of entries n is even, it is the arithmetic mean
of the two central numbers.

2
 • Mode: The value or categorical entry displaying the highest frequency within the
dataset.

SAT Structural Strategy: Skewness & Outliers

An outlier is an extreme data point located an abnormal distance from the re-
maining values.

• Outliers pull the Mean aggressively toward the extreme tail.

• Outliers rarely shift the Median, leaving it spatially robust. Hence, for
skewed metrics (like real estate prices or national wealth assets), the median
provides a more realistic midpoint.

Frequency
Mode
Median
Mean

Value
Right-Skewed Model (Positive Skew): Mean > Median > Mode

5 Advanced Dispersion & Grouped Data

(Advanced Extension: Vietnamese High School Curricu-
lum)
While the SAT evaluates structural range (Max−Min) and qualitative standard deviation
profiles, the advanced Vietnamese curriculum (Toán 10, Toán 11) demands computational
mastery over grouped variance profiles.

5.1 1. Population Variance and Standard Deviation

Standard Deviation (σ) measures the precise average distance between individual
observations and the system mean.

• Variance (σ 2 ): The absolute arithmetic mean of all squared deviations from the
system average.

• Mathematical Formula:
sP
i=1 (xi − x̄) i=1 (xi
Pn 2 n
− x̄)2
σ2 = =⇒ σ = (3)
n n

• Conceptual SAT Operational Axiom: If Dataset A is heavily clustered around its

midpoint while Dataset B displays a widely dispersed range, Dataset B contains a
higher standard deviation, irrespective of their absolute mean positions.

3
5.2 2. Grouped Continuous Data Statistics (S liu ghép nhóm -
Toán 11)
When data is classified in continuous closed intervals [ui ; ui+1 ) due to measurement con-
straints, summary analysis uses the interval midpoint (ci = ui +u2 i+1 ) to approximate
population metrics:
Pk k
" #
i=1 ni ci
x̄ = where ni = frequency of class i, and N = (4)
X
ni
N i=1

5.3 3. Five-Number Summary and Box Plot Formats

A Box Plot structural model projects five critical distribution benchmarks: Minimum,
Q1 (First Quartile / 25th percentile), Q2 (Median / 50th percentile), Q3 (Third Quartile
/ 75th percentile), and Maximum.
• Interquartile Range (IQR): IQR = Q3 − Q1 (Contains the middle 50% of the
data volume).

Q1 Q2 (Med) Q3
Min Max

1 2 3 4 5 6 7 8 9

6 Bivariate Inferences & Scatter Plots

A scatter plot visualizes pairs of continuous variable markers across an experimental
cohort to uncover structural dependencies or directional patterns.
• Line of Best Fit: A calculated linear
regression trend equation minimizing total y (Dependent)
point coordinate distance.

• Functional Form: y = mx + b

• Slope (m): Explains the change in output

y predicted for each **single-unit step in-
crease** in variable x. Line of Best Fit
• y-intercept (b): The baseline value pre- x (Independent)
dicted for y when input x = 0.
SAT Trap Strategy: Observation vs. Prediction Models

Always separate an actual data point (represented as an empirical dot) from a

predicted outcome value (modeled cleanly on the continuous line of best fit).
If a question evaluates real empirical variance, calculate from the dot. If it queries
expected trends, read coordinates from the line.

Residual Value = Actual Value − Predicted Value

4
7 Probability and Counting Principles
Probability quantifies structural uncertainty by measuring event outcomes against a base-
line sample space.
Count of Favorable Outcomes
P (A) = (5)
Total Cardinality of the Sample Space (Ω)

7.1 1. Two-Way Categorical Matrix tables & Conditional Prob-

ability
The SAT checks conditional probability distributions from structured multi-variable rows.

Passed Assessment Failed Assessment Total

Completed Prep 80 5 85
No Prep Course 15 20 35
Total 95 25 120

• Marginal Probability Event: Find the likelihood that a random choice candidate
passed the assessment: 120
95
.

• Conditional Probability Event: Given that the chosen candidate had no prep
course, find the likelihood that they passed: Here, the sample universe isolates to
the second row. Answer: 35 15
.

7.2 2. Advanced Combinatorics (Extension: Vietnamese Grade

10 Math)
When calculating intricate compound sample spaces, use structured counting rules:

• Permutations (Akn ): Set configurations where selection order matters (e.g., as-
signment roles).
n!
Akn = (6)
(n − k)!

• Combinations (Cnk ): Group arrangements where internal order is irrelevant (e.g.,

drafting committees). !
n n!
Cn =
k
= (7)
k k!(n − k)!

5
8 Pro SAT Data Analysis Tips
Key Design Strategies for a Perfect 800
1. Control Margin of Error: The margin of error bounds random sample
selection noise. It can only be compressed by expanding the absolute
sample population size (n). It is not an indicator of systematic study
design mistakes!

2. Random Selection vs. Random Assignment:

• Random Selection empowers an investigator to project localized trends

out onto the **entire parent population**.
• Random Assignment systematically blocks lurking variables, verifying
direct **cause-and-effect** conclusions.

3. Examine Axis Scaling Metrics: Verify whether coordinates scale in dis-

tinct thousands, millions, or logarithmic units prior to computing numeric
calculations.

End of Syllabus. Ready for Student Distribution.