
summarize categorical data in a two-way frequency table, interpret joint, marginal and conditional relative frequencies in context, and decide whether the data suggest an association.
Instruction
Some data are numbers you can average, like heights or test scores. Other data are categories: grade level, yes or no, rural or urban. When each person in a survey answers two categorical questions, a two-way frequency table shows every combination at once. This lesson is about reading such a table well: turning its counts into relative frequencies, saying what each one means in context, and judging whether the two variables seem to be associated.
1. The table. In an illustrative survey, 200 Missouri high school students, 100 ninth graders and 100 twelfth graders, were asked whether they have a part-time job.
| Job | No job | Total | |
|---|---|---|---|
| Grade 9 | 18 | 82 | 100 |
| Grade 12 | 54 | 46 | 100 |
| Total | 72 | 128 | 200 |
The inside cells are joint counts, grade and job status together. The totals in the margins are marginal counts.
2. Three kinds of relative frequency. A relative frequency is a count divided by a total, and the total you choose decides the question you answer.
- Joint: 54 ÷ 200 = 0.27. Of all students surveyed, 27% are twelfth graders with a job.
- Marginal: 72 ÷ 200 = 0.36. Of all students surveyed, 36% have a job.
- Conditional: 54 ÷ 100 = 0.54. Of the twelfth graders, 54% have a job. Compare 54 ÷ 72 = 0.75: of the students with a job, 75% are twelfth graders. Same cell, different denominator, different meaning.
Always name the group in words, as in “of the twelfth graders” or “of students with a job.” Most errors in reading two-way tables come from dividing by the wrong total.
3. Looking for an association. Two categorical variables are associated when the conditional relative frequencies differ from group to group. Compute row percentages: 18% of ninth graders have a job, but 54% of twelfth graders do. The rates differ by a wide margin, so the data suggest an association between grade level and having a job. Had both rows shown the same rate, the table would show no association.
4. Why counts mislead. A second illustrative survey asked 400 students whether they ride a school bus. Of 150 rural students, 60 ride; of 250 urban students, 75 ride. More urban students ride, 75 against 60, but the rural rate is higher: 60 ÷ 150 = 40% against 75 ÷ 250 = 30%. When groups differ in size, compare conditional relative frequencies, never raw counts.
5. Association is not cause. The job table does not show that being a senior causes students to work. Older students may have more free time, a driver’s license or the legal age for more kinds of jobs. A two-way table can reveal a pattern worth explaining; it cannot by itself explain it. When you write a claim, say “the data suggest an association,” give the rates that support it, and name one other factor that could account for it.
📖 Read for this lesson
The reading opens over the lesson; close it to come back to exactly where you were. The checks below draw on it.
Formative check
Work through these before moving on. They are not graded — they tell you whether the standard below has landed yet.
Missouri standard: A1.DS.A.4a
In the job survey, what percent of twelfth graders have a part-time job?
Missouri standard: A1.DS.A.4a
In the job survey, 18 ÷ 200 = 0.09. What does 0.09 mean?
Missouri standard: A1.DS.A.4a
In the bus survey, % of the rural students ride a school bus.
Missouri standard: A1.DS.A.4b
Because 75 urban students ride the bus and only 60 rural students do, urban students in the survey are more likely to ride the bus.
Missouri standard: A1.DS.A.4b
Which survey result shows no association between the two variables?
How confident are you that you can read a two-way table, interpret joint, marginal and conditional relative frequencies, and judge an association?
Practice
Work these on paper or in your notebook, then open Check your answer. Aim for all of Fluency and Application; try at least one Challenge.
Printable version: this unit’s practice workbook (PDF)
Fluency
Build speed and accuracy with the core skill.
- Homework device survey (160 students): grade 9, laptop 48 and phone 32; grade 12, laptop 60 and phone 20. Find the joint relative frequency of twelfth graders who use a laptop.
Check your answer
Answer: 60 ÷ 160 = 0.375 - In the homework device survey, find the marginal relative frequency of laptop users.
Check your answer
Answer: 108 ÷ 160 = 0.675 - In the homework device survey, what percent of ninth graders use a phone?
Check your answer
Answer: 32 ÷ 80 = 40% - In the homework device survey, what percent of phone users are ninth graders?
Check your answer
Answer: 32 ÷ 52 ≈ 61.5% - Complete the table: grade 9 has 30 yes and a row total of 75; grade 12 has 45 yes and 30 no. Find the missing count and every total.
Check your answer
Answer: Grade 9 no = 45; grade 12 total 75; yes total 75; no total 75; grand total 150. - Breakfast survey: 63 of 90 students with a sibling and 42 of 60 without one eat breakfast. Give the row percentages.
Check your answer
Answer: Sibling: 70% yes, 30% no. No sibling: 70% yes, 30% no.
Application
Use the skill in context. Show your reasoning.
- Do the homework device data suggest an association between grade and device? Support your answer with conditional relative frequencies.
Check your answer
Answer: Yes: 60% of ninth graders (48 ÷ 80) and 75% of twelfth graders (60 ÷ 80) use a laptop, so the rates differ. - Do the breakfast data suggest an association between having a sibling and eating breakfast?
Check your answer
Answer: No: the rate is 70% in both groups. - A classmate says urban students are more likely to ride the bus because 75 urban and 60 rural students ride. Of 250 urban and 150 rural students surveyed, explain the error.
Check your answer
Answer: The groups differ in size, so compare rates: 40% of rural (60 ÷ 150) and 30% of urban (75 ÷ 250) students ride. Rural students are more likely to ride. - Of 120 students, 50 play a school sport. 30 of the sport players and 20 of the other students report 8 or more hours of sleep. Build the two-way table, compare the rates and state a careful conclusion.
Check your answer
Answer: Sport: 30 yes, 20 no (60%). No sport: 20 yes, 50 no (about 28.6%). The data suggest an association between playing a sport and reporting more sleep, but they do not show that sports cause it; schedules or age could account for it.
Challenge
Stretch problems. Expect to think before you write.
- Design a two-way table for 200 students, 100 in each grade, in which 30% of all students say yes and there is no association between grade and answer. Then change two cells so that an association appears while the column totals stay the same.
Check your answer
Answer: No association: 30 yes and 70 no in each grade. With an association, for example: grade 9 has 20 yes and 80 no, grade 12 has 40 yes and 60 no (still 60 yes and 140 no).
Review
Keep earlier skills sharp.
- For the data 8, 9, 11, 13, 17, 34, 36: find the mean, median, range and interquartile range. (Find the quartiles as the medians of the lower and upper halves, leaving out the overall median when the count is odd.)
Check your answer
Answer: mean = 18.29; median = 13.0; range = 28; IQR = 25.0 (Q1 = 9.0, Q3 = 34.0) - An arithmetic sequence has first term −6 and common difference 4. Write an explicit formula, then find term 16 and the sum of the first 16 terms.
Check your answer
Answer: an = −6 + (4)(n − 1); a16 = 54; S16 = 384
Four Student Surveys in Two-Way Tables
Constructed teaching data written for this course; the surveys are illustrative, not real results · SLS original (constructed teaching data)
Why this reading. The survey brief asks you to turn a table into relative frequencies and to judge an association. These four tables are the data: two are worked in the lesson, and two are yours.
Survey 1. Part-time jobs by grade (200 students)
| Job | No job | Total | |
|---|---|---|---|
| Grade 9 | 18 | 82 | 100 |
| Grade 12 | 54 | 46 | 100 |
| Total | 72 | 128 | 200 |
Row percentages: 18% of ninth graders and 54% of twelfth graders have a job.
Survey 2. School bus riders, rural and urban (400 students)
| Rides | Does not ride | Total | |
|---|---|---|---|
| Rural | 60 | 90 | 150 |
| Urban | 75 | 175 | 250 |
| Total | 135 | 265 | 400 |
Row percentages: 40% of rural and 30% of urban students ride.
Survey 3. Main homework device by grade (160 students)
| Laptop | Phone | Total | |
|---|---|---|---|
| Grade 9 | 48 | 32 | 80 |
| Grade 12 | 60 | 20 | 80 |
| Total | 108 | 52 | 160 |
Survey 4. Breakfast and siblings (150 students)
| Eats breakfast | Skips breakfast | Total | |
|---|---|---|---|
| Has a sibling | 63 | 27 | 90 |
| No sibling | 42 | 18 | 60 |
| Total | 105 | 45 | 150 |
Tasks
- For Surveys 3 and 4, compute one joint, one marginal and two conditional relative frequencies, and write each as a sentence that names the group.
- Convert each survey to row percentages and decide whether it suggests an association.
- For each survey that does, name one other factor that could explain the pattern, so your claim stops short of cause.