SAT Two-Way Tables: Rows, Columns, Conditional Percentages, and Probability Traps

A two-way table organizes data for two categorical variables at the same time.

The SAT can use one table to ask about:

• total counts;
• joint frequencies;
• row or column percentages;
• ordinary probability;
• conditional probability;
• missing values;
• comparisons between groups.

The table itself is usually easy to read. The harder part is deciding which total belongs in the denominator.

Use this rule:

• probability of A and B → divide the overlap by the grand total;
• probability of A given B → divide the overlap by the total for B;
• percentage within a row/column → divide by that row/column total.

Why two-way tables are explicitly relevant to the SAT

College Board's Problem-Solving and Data Analysis domain includes probability and conditional probability, and current test specifications note that students may use two-way tables to find relative frequencies and probabilities.

Two-way tables also connect percentages and data interpretation, so a single question may require several skills at once.

For the broader probability topic, see SAT Probability and Statistics. For "given that" questions, see SAT Conditional Probability.

Anatomy of a two-way table

Suppose a school surveys 200 students:

The interior cells are joint frequencies:

• Grade 11 and plays sport: $50$
• Grade 12 and does not: $50.$

The rightmost column contains row totals:

$$80,\ 120.$$

The bottom row contains column totals:

$$120,\ 80.$$

Grand total:

$$200.$$

Joint frequency vs marginal frequency

Joint frequency

Count satisfying two categories simultaneously.

Example:

$$70$$

students are both Grade 12 and play a sport.

Marginal frequency

A row/column total that summarizes one category.

Example:

$$120$$

students are Grade 12.

The term "marginal" comes from totals appearing at the margins of the table.

You do not need jargon to solve the question, but knowing the distinction helps organize the denominator.

Worked example 1: ordinary probability

Using the table above, what is the probability a randomly selected student is a Grade 12 student who plays a sport?

The target is:

Grade 12 and sport.

Count:

$$70.$$

No condition restricts the population, so denominator is the grand total:

$$200.$$ $$P=\frac{70}{200}=0.35.$$ $$\boxed{35\%}$$

Worked example 2: conditional probability

What is the probability a student plays a sport given that the student is in Grade 12?

Now restrict to Grade 12.

Denominator:

$$120.$$

Grade 12 sport players:

$$70.$$ $$P=\frac{70}{120}=\frac7{12}\approx0.5833.$$ $$\boxed{\frac7{12}\text{ or about }58.3\%}$$

Same numerator as before; different denominator.

Worked example 3: reverse the condition

What is the probability a sport-playing student is in Grade 12?

Condition:

plays sport.

Denominator:

$$120.$$

Grade 12 sport players:

$$70.$$ $$P=\frac{70}{120}=\frac7{12}.$$

In this particular table it happens to equal the prior probability because both relevant marginal totals equal 120.

That is coincidence.

In general:

$$P(A\mid B)\ne P(B\mid A).$$

Always choose the denominator from the condition rather than assuming symmetry.

Row percentages

A row percentage compares cells within one row.

Grade 11 sport percentage:

$$\frac{50}{80}=62.5\%.$$

Grade 12 sport percentage:

$$\frac{70}{120}\approx58.3\%.$$

So even though Grade 12 has more sport players in raw count, Grade 11 has the higher within-grade percentage.

This raw-count vs percentage distinction is a common data trap.

Worked example 4: count vs proportion

Suppose:

Group B has more supporters:

$$90>60.$$

But support rates:

Group A:

$$60/100=60\%.$$

Group B:

$$90/200=45\%.$$

So Group A has the higher proportion of supporters.

The question wording must decide whether you compare counts or percentages.

Column percentages

Suppose you ask:

Of all students who play a sport, what percentage are Grade 11?

Sport column total:

$$120.$$

Grade 11 sport players:

$$50.$$ $$\frac{50}{120} = \frac5{12} \approx41.7\%.$$ $$\boxed{41.7\%}$$

The denominator is the column total because the question starts with the sport-playing subgroup.

"Of those..." language

Conditional percentages often avoid formal probability wording.

Examples:

• "Of students in Grade 12, what percent play a sport?"
• "Among customers who bought product A, what fraction also bought B?"
• "For respondents under age 30, what proportion chose option C?"

All mean:

Restrict to the named group first.

Then compute within that group.

Missing table entries

Two-way tables often include totals that let you recover missing cells.

Example:

From Group A:

$$28+x=70.$$ $$\boxed{x=42}$$

Check No column:

$$42+38=80.$$

Consistent.

Worked example 5: recover a missing total

A-column total:

$$24+36=60.$$ $$\boxed{60}$$

Also:

$$60+60=120.$$

Use both row/column consistency checks when possible.

Convert counts to relative-frequency table

Take:

Because the grand total is 100, overall relative frequencies are numerically the same as percentages:

• X and A: $20\%.$
• Y and B: $40\%.$

But a row-relative-frequency table would divide by row totals instead.

For X:

$$20/50=40\%,\quad30/50=60\%.$$

A table can be normalized in different ways. Read labels carefully.

Worked example 6: relative frequency from a non-100 total

Out of 250 respondents, 75 are in Group A and choose option X.

Joint relative frequency:

$$75/250=0.30.$$ $$\boxed{30\%}$$

If Group A contains 100 respondents total, the conditional percentage choosing X is:

$$75/100=75\%.$$

Same cell. Different denominator.

Two-way tables and independence

Two categorical variables are independent when the distribution of one is the same across levels of the other.

Example:

Overall:

$$P(A)=40\%.$$

Within B:

$$P(A\mid B)=20/50=40\%.$$

Within not B:

$$20/50=40\%.$$

The A percentage does not change across groups.

This supports independence.

Worked example 7: identify association

Within Group A:

$$70\%.$$

Within Group B:

$$40\%.$$

The distributions differ substantially.

So group membership and choosing X are associated in this sample.

That does not prove one variable causes the other. Study design matters.

Two-way tables and "at least one"

Suppose categories overlap and the table gives two binary traits.

To find probability of A or B:

$$P(A\cup B) = P(A)+P(B)-P(A\cap B).$$

In counts:

$$\#(A\cup B) = \#A+\#B-\#(A\cap B).$$

You subtract the overlap once because adding the two marginal totals counts it twice.

Worked example 8: union from counts

Out of 200 people:

• 90 have A;
• 80 have B;
• 30 have both.

Number with A or B:

$$90+80-30=140.$$

Probability:

$$140/200=70\%.$$ $$\boxed{70\%}$$

Complements from a table

If a row total is 120 and 70 satisfy event A, then:

$$120-70=50$$

do not.

Conditional probability of not A within that row:

$$50/120.$$

Or simply:

$$1-\frac{70}{120}.$$

Use whichever is faster.

Worked example 9: compare conditional percentages

A survey table:

Under 30 use rate:

$$84/120=70\%.$$

30+ use rate:

$$90/180=50\%.$$

Difference:

20 percentage points

Relative difference compared with 50%:

$$\frac{70-50}{50}=40\%.$$

If asked about percentage points vs relative percent, see SAT Percentages and Percent Increase vs Percentage Points.

Tables in SAT Reading and Writing vs SAT Math

Math two-way tables test quantitative reasoning.

Reading and Writing data questions may also use tables or graphs, but the task there is usually to choose a statement that accurately supports a textual claim rather than compute a full probability.

For that different skill, see SAT Data Evidence Questions.

Common trap 1: use the grand total for every percentage

Conditional questions use subgroup totals.

Common trap 2: read the wrong row or column

Translate the words into categories before calculating.

Common trap 3: compare raw counts instead of rates

Groups of different sizes require percentage/proportion comparison if the question asks prevalence.

Common trap 4: reverse the condition

"Among A, how many B?" uses A as denominator.

Common trap 5: double-count an overlap

For A or B, subtract A-and-B once.

Common trap 6: assume a table shows causation

A table describes association unless study design supports stronger inference.

Common trap 7: ignore labels on a relative-frequency table

Values may be overall, row-normalized, or column-normalized.

Common trap 8: calculate more totals than needed

For a conditional probability, you may need only one row or column.

A reliable two-way-table routine

1. Identify the two categorical variables.
2. Mark the exact cell relevant to the target.
3. Decide whether denominator is:
   • grand total;
   • row total;
   • column total.
4. Compute fraction/decimal/percent.
5. Check whether the question asks count or proportion.
6. Use totals to verify missing cells if needed.

SAT-style practice

Use:

Question 1

Probability of Group A and Yes:

$$36/150=0.24.$$ $$\boxed{24\%}$$

Question 2

Probability of Yes given Group A:

$$36/60=0.60.$$ $$\boxed{60\%}$$

Question 3

Probability of Group A given Yes:

$$36/60=0.60.$$ $$\boxed{60\%}$$

Again equal here by coincidence because both margins are 60.

Question 4

What percent of Group B answered Yes?

$$24/90 = \frac4{15} \approx26.7\%.$$ $$\boxed{26.7\%}$$

Question 5

Which group has the higher Yes rate?

Group A:

$$60\%.$$

Group B:

$$26.7\%.$$

Group A

FAQ: SAT two-way tables

What is a two-way table?

A table that cross-classifies observations by two categorical variables.

What is a joint frequency?

An interior cell count satisfying both its row and column category.

What is a marginal frequency?

A row or column total.

How do I find conditional probability from a two-way table?

Use the relevant overlap as numerator and the conditioned row/column total as denominator.

How do I compare groups of different sizes?

Compare within-group proportions or percentages, not raw counts, when the question asks relative prevalence.

Can a two-way table show independence?

Yes. If conditional proportions match the overall or cross-group proportions, the variables may be independent.

Does association in a two-way table prove causation?

No. Causation depends on study design, particularly random assignment in experiments.

Are two-way tables explicitly part of SAT probability?

Yes. College Board's specifications include two-way tables among representations used for probability and conditional probability.

How to practice this on MastaPrep

The quickest way to lock in a skill like this is to practice it on its own, then again a few days later, mixed in with other questions. That is how MastaPrep is built. A placement check finds your weakest SAT Math areas, the study plan targets them, every question has a worked explanation, and missed questions come back until you get them right without help.

See how MastaPrep works or try free SAT practice questions. The free plan needs no credit card.

Related MastaPrep guides

• SAT Conditional Probability
• SAT Probability and Statistics
• SAT Percentages
• SAT Data Evidence Questions
• SAT Ratios, Rates, and Unit Conversions
• SAT Sampling Bias

Final takeaway

A two-way table is mostly a denominator test.

For any question, ask:

Am I looking at everyone, one row, or one column?

Then choose:

$$\boxed{\text{target count}/\text{relevant total}}$$

The arithmetic is usually simple once the comparison group is correct.

---

Official SAT references:
https://satsuite.collegeboard.org/sat/whats-on-the-test/math/types/problem-solving
https://satsuite.collegeboard.org/practice/student-question-bank/math

Trademark note: SAT® is a registered trademark of College Board. MastaPrep is independent and is not affiliated with or endorsed by College Board.