On this concept: Formula

Not sure which lesson to do first? Find your level and get a plan →

Statistics · Grade 8-12

Conditional Relative Frequency

Conditional relative frequency is the proportion of cases in a chosen group of a two-way table that also fall in another category — the cell count divided by that group's row or column total, not the grand total.

conditional relative frequency=cell countrelevant row or column total

Quick Answer

Conditional relative frequency is the proportion of cases in a chosen group of a two-way table that also fall in another category — the cell count divided by that group's row or column total, not the grand total. Use it when a question narrows the group first, as in "of the students who play sports" or "among the vaccinated", and then asks what share of that group has some other feature. If 30 students play sports and 18 of them have jobs, the answer is 18/30=0.60; the share 18/200=0.09 of all 200 students answers a different question.

Why This Matters

Conditional relative frequencies are how you test for an association in categorical data: if 60% of athletes have jobs but only 20% of non-athletes do, the two variables are linked; if both groups showed 26%, they would not be. Divide by the wrong total and you either hide a real difference or invent one, and many misleading headlines come from exactly that slip.

Intuitive Explanation

Picture the two-way table as a building with rooms. A joint relative frequency compares one room with the whole building: of all 200 students, 18 play sports and have a job, so 18/200=9%. A conditional relative frequency first walks into one wing and closes the door behind it. Inside the sports wing there are only 30 students, and 18 of them have jobs, so the share is 18/30=60%. Same cell, different whole.

Doing this for each wing makes the groups comparable. In the non-sports wing, 34 of 170 students have jobs, which is 20%. Now the comparison is fair, 60% against 20%, even though the non-sports group has more job-holders by raw count (34 against 18). If the conditional shares were equal in every wing, the two variables would show no association; the bigger the gap, the stronger the link.

Which wing you enter matters. "Of the athletes, what share have jobs?" divides by the sports row, 30. "Of the students with jobs, what share are athletes?" divides by the job column, 52, and gives 18/52≈35%. Both are conditional relative frequencies, but they answer different questions, and swapping them is one of the most common errors in reading data. The words right after "of" or "given that" name the denominator.

Think of it like this. A conditional relative frequency works like a zoom lens on the table: you frame one row or one column, and that frame becomes the whole picture, so its total goes underneath. Zoom on a different row or column and the same cell becomes a different percentage.

Core idea. The denominator is the group you condition on: pick one row or column, divide its cells by that group's own total, and compare those shares across groups to see whether two categorical variables are associated.


When to Use

Use it when a two-way table comes with a question that narrows the group first: of those who, among, given that, or "what percent of the girls". Name the group, divide by its row or column total, and compare across groups. If the question is about everyone, such as "what percent of all students are girls who play chess", divide by the grand total instead (a joint relative frequency); if it describes a single random draw in terms of chance, use conditional probability.

Tip. Ask: Does the question narrow to one row or column of a two-way table first, then ask for a share of that group?

How to Recognize It

Before dividing, find the group the question picks out and decide which total belongs on the bottom of the fraction.

  1. Does the question name a group first, as in 'of the 8th graders' or 'among students who ride the bus'?

    Yes means that group's row or column total is the denominator. If it asks for a share of everyone surveyed, divide by the grand total, which is a joint share from plain Relative Frequency.

  2. Would swapping the two categories in the question change which total you divide by?

    Yes: 'of the bus riders, what share are 8th graders' divides by the bus column, while 'of the 8th graders, what share ride the bus' divides by the 8th-grade row, and the answers usually differ.

  3. Are you working from counts that were actually observed, rather than probabilities given in advance?

    Yes keeps you in Conditional Relative Frequency. If the problem states P(A∩B)=0.12 and P(B)=0.4 and asks for P(A∣B), use Conditional Probability; the division is the same.

  4. Does the question want a share, not just how many are in one cell?

    Yes needs the division. A question like 'how many 8th graders ride the bus' is answered by reading one cell of the Two-Way Tables grid, with no denominator at all.

Conditional Relative Frequency vs Relative Frequency vs Two-Way Tables vs Conditional Probability

All four live around the same grid of counts and often use the very same cell. What separates them is the denominator: one group's total, the grand total, no division at all, or a probability.

Conditional Relative Frequency

Meaning
A conditional relative frequency is a cell count divided by the total of the one row or column the question conditions on.
Key test
The bottom of the fraction is a single group's total.
Formula
cell countrow or column total
Example
52 of the 80 eighth graders buy lunch, so the conditional relative frequency is 52/80=0.65.

Relative Frequency

Meaning
A relative frequency divides a count by the grand total; for a single cell this is called a joint relative frequency.
Key test
Divides by everyone in the table, not by one group.
Formula
countgrand total
Example
52 of all 200 students are eighth graders who buy lunch: 52/200=0.26.

Two-Way Tables

Meaning
A two-way table is the grid of counts for two categorical variables, with row, column and grand totals in the margins.
Key test
Reading or filling in counts; nothing is divided yet.
Formula
row total=∑cells in the row
Example
Rows for 7th and 8th grade, columns for buys lunch and brings lunch, 200 students in all.

Conditional Probability

Meaning
Conditional probability is the chance of an event given that another has happened, usually worked from stated probabilities.
Key test
Starts from probabilities such as P(A∩B), not observed counts.
Formula
P(A∣B)=P(A∩B)P(B)
Example
If P(8th grade and buys)=0.26 and P(8th grade)=0.40, then P(buys∣8th grade)=0.65.

Formula & Notation

conditional relative frequency=cell countrelevant row or column total
In a two-way table, conditional relative frequencies normalize counts within a selected row or column so categories can be compared fairly across groups of different sizes.

How to read it: Joint relative frequency uses the grand total in the denominator. Marginal relative frequency divides a row or column total by the grand total. Conditional relative frequency divides a cell count by its row or column total.

Worked Examples

Example 1 — Do athletes hold jobs more often?

Easy

Of 200 students, 30 play a sport and 170 do not. Among the athletes, 18 have a part-time job; among the non-athletes, 34 do. Is having a job more common among athletes?
Show the steps and the answer
  1. Lay out the rows: sports 18 job, 30−18=12 no job; no sports 34 job, 170−34=136 no job.

    The question compares two groups, athletes and non-athletes, so each group is a row with its own total.

  2. Condition on athletes: 1830=0.60.

    "Among the athletes" makes the sports row total, 30, the denominator.

  3. Condition on non-athletes: 34170=0.20.

    Each group is divided by its own total, so the comparison is not distorted by the groups' sizes.

  4. Compare: 60% against 20%, while the raw counts were 18 against 34.

    By count, non-athletes look more likely to work; by share, athletes are three times as likely.

Answer: Yes. 60% of athletes have a job, compared with 20% of non-athletes, so the two variables are associated in this group.

Takeaway: Compare groups with conditional shares inside each group, never with the raw cell counts.

Example 2 — Same cell, three denominators

Standard

Use the same table: 18 students play a sport and have a job, 52 students have a job in all, 30 play a sport, and there are 200 students. (a) Of the students with jobs, what fraction play a sport? (b) What fraction of all students are athletes with jobs?
Show the steps and the answer
  1. (a) Condition on job-holders: 1852≈0.346.

    "Of the students with jobs" puts the job column total, 52, underneath, not the sports row total, 30.

  2. (b) Use the grand total: 18200=0.09.

    "Of all students" does not narrow the group, so this is a joint relative frequency, not a conditional one.

  3. Set the three side by side: 18/30=0.60, 18/52≈0.35, 18/200=0.09.

    One cell gives three different numbers; only the wording of the question picks the right one.

Answer: (a) About 0.35 of job-holders play a sport. (b) 0.09 of all students are athletes with jobs. Neither equals 0.60, the share of athletes who have jobs.

Takeaway: Reversing the condition or dropping it changes the denominator, and with it the answer.

Example 3 — Find the mistake: the grand-total trap

Application

A survey of 200 students: of 120 who ate breakfast, 24 felt tired by 10 a.m.; of 80 who skipped breakfast, 48 felt tired. Jordan writes: "Of breakfast eaters, 24/200=12% were tired; of skippers, 48/200=24% were tired." Find the mistake and correct it.
Show the steps and the answer
  1. Spot the denominator: Jordan divided both cells by 200, the grand total.

    Dividing by 200 gives the share of all students who are, say, breakfast eaters and tired, a joint relative frequency.

  2. Breakfast eaters: 24120=0.20.

    "Of breakfast eaters" narrows the group to 120 students, so 120 goes underneath.

  3. Skippers: 4880=0.60.

    The skippers' own total is 80.

  4. Compare: 20% against 60%, a threefold gap; Jordan's numbers showed only a twofold gap.

    Using the grand total shrinks both shares by different amounts, because the two groups have different sizes, so it distorts the comparison.

Answer: 20% of breakfast eaters and 60% of skippers felt tired by 10 a.m.

Takeaway: When the question says "of" a group, that group's total, not the grand total, is the denominator.

Where You'll See It

  • Snack choice within just the 6th graders

    From a grade-by-snack table, 'of the 6th graders, what percent chose chips?' divides the 6th-grade-chips cell by the 6th-grade row total — not by the whole school.

  • App use among phone owners only

    To find what fraction of phone owners use a certain app, you divide the (owns phone AND uses app) cell by the phone-owner total, restricting attention to that one group.

  • Test results among the vaccinated group

    'Among vaccinated students, what proportion got sick?' uses the vaccinated row total as the denominator, letting you compare it fairly to the unvaccinated group's conditional rate.

Common Mistakes

  1. The slip: Divide the cell count by the grand total to get a conditional relative frequency.

    Why it's tempting: The grand total is the most obvious 'total' on the table.

    Instead: Use the relevant row OR column total (the group you're conditioning on), not the grand total. The grand total gives a joint relative frequency, a different quantity.

  2. The slip: Conditioning on row vs. column gives the same percentage.

    Why it's tempting: The same cell appears in both, so the result seems fixed.

    Instead: The denominator differs: 'percent of A's that are B' (row) versus 'percent of B's that are A' (column) are generally different numbers; choose the one the question asks for.

  3. The slip: Conditional relative frequency and conditional probability P(A|B) are unrelated.

    Why it's tempting: One comes from a data table and the other from probability notation, so they feel like separate topics.

    Instead: A conditional relative frequency IS the data-based estimate of a conditional probability: cell count over the conditioning group's total mirrors P(A∣B).


Mini Practice

Try these on your own. Tap Reveal when you want to check.

  1. Of 80 students, 50 are in band and 30 are not. 20 band members and 6 non-band students take art. Find the conditional relative frequency of taking art among band members and among non-band students.

    Hint: Use the band total for the first and the non-band total for the second.

  2. A two-way table shows that 30 students play sports and 18 of those also have jobs. Find the conditional relative frequency of having a job, given that a student plays sports.

    Hint: Which group's total does "given that a student plays sports" fix as the denominator?

  3. A survey of 200 adults: 90 prefer coffee, and 54 of them are over 40; 110 prefer tea, and 33 of them are over 40. Among people over 40, what fraction prefer coffee?

    Hint: First find how many people are over 40 in total.

  4. 60 of 80 phone owners use social media daily, compared with 12 of 40 people without a phone. Use conditional relative frequencies to decide whether the table shows an association.

    Hint: Find the daily-use share inside each group and compare them.

  5. Of 200 people surveyed, 36 are both employed and over 40. Is "36 out of 200" a conditional relative frequency?

    Hint: Is the share taken from the grand total or from one row or column?

  6. A class of 50 has 25 students who walk to school, and 10 of the walkers are in Grade 7. Zara writes, "Of the walkers, 1050=20% are in Grade 7." Correct her work.

    Hint: "Of the walkers" tells you which total to divide by.

Want the full set?

50 practice questions for this concept — free to try, every one with a complete worked solution showing the why, not just the answer.

Frequently Asked Questions

Can a group look better overall but worse in every subgroup?

Yes, and it is called Simpson's paradox. Hospital A saves 70 of 100 severe cases (70%) and 19 of 20 mild ones (95%). Hospital B saves 12 of 20 severe cases (60%) and 90 of 100 mild ones (90%). A does better in both groups, yet overall A saves 89 of 120 (74%) and B saves 102 of 120 (85%), because A treats far more severe cases. Check whether a third variable splits the table before comparing.

Why can a medical test that catches 90% of cases still give so many false alarms?

Because two different conditional relative frequencies come from the same table. In 1,000 people, 20 have a disease and the test flags 18 of them, so 18/20 = 90% of sick people test positive. But 49 of the 980 healthy people also test positive. Of all 67 positive results, only 18/67, about 27%, belong to sick people. Same cell, different group, very different message.

Why do the conditional percents across a row add to 100% but the ones down a column do not?

When you condition on rows, every percent in a row shares one denominator, that row's total, so together they cover the whole row: among 8th graders, 65% buy lunch and 35% bring it. Going down a column, each percent came from a different row total, so they describe different groups and need not add to anything. If you condition on columns instead, the columns are the ones that add to 100%.

Can I find a conditional relative frequency if I only have percents of the whole table?

Yes: divide the joint share by the share of the group you condition on. If 26% of all students are 8th graders who buy lunch and 40% of all students are 8th graders, then the share of 8th graders who buy lunch is 0.26/0.40=0.650.26 / 0.40 = 0.65. This is the same step as P(A∣B)=P(A∩B)/P(B)P(A \mid B) = P(A \cap B) / P(B), written with data instead of probabilities.

Learning Path

Conditional Relative Frequency

You are here

Build on Two-Way Tables (cells and margins) and Relative Frequency (count over total). This concept changes the total: you divide by a single row or column, so groups of different sizes can be compared and associations between categories show up. Next, Conditional Probability writes the same idea as P(A∣B), and Correlation measures association when both variables are numerical.

See Also