When a study records two categorical variables for the same individuals, you can organize the results in a two-way table and graph them with side-by-side bar graphs, segmented bar graphs, or mosaic plots. Those displays help you compare conditional distributions and decide whether the variables appear associated.
Why This Matters for the AP Statistics Exam
This topic is the first place in the revised course where you analyze the relationship between two variables instead of just one. On the AP Statistics exam, you may need to read a two-way table, compare conditional proportions, or explain whether a display suggests an association between variables. This work also sets up later probability and inference topics that rely on the same table structure.
On both multiple-choice and free-response questions, you should expect to move back and forth between counts, proportions, and context. Strong responses compare the right distributions instead of just quoting raw counts.
Key Takeaways
- A two-way table, also called a contingency table, summarizes two categorical variables using rows and columns.
- A joint relative frequency is one cell divided by the grand total.
- A conditional relative frequency uses a row total or column total and is usually what you compare when deciding whether variables are associated.
- Side-by-side bar graphs, segmented bar graphs, and mosaic plots all display one categorical variable broken down by another.
- Mosaic plots use area to represent joint relative frequencies.
- Association is not the same as causation.
Starting with the Investigative Question
With one categorical variable, you describe a distribution. With two categorical variables, you usually ask whether the distribution of one variable changes across the categories of the other.
Examples of investigative questions include:
- Does sleep amount differ across grade levels?
- Is preference for a study method related to whether a student plays a sport?
- Are students in different programs equally likely to pass an exam?
Those questions do not ask for cause and effect. They ask whether the variables appear related.
Two-Way Tables
A two-way table, also called a contingency table, has rows and columns that match the categories of the two variables. Each cell holds the count or proportion of data points that fall into that combination of categories. This makes it easy to see how individuals are distributed across both variables at once.
Joint and Conditional Relative Frequencies
You can also fill a two-way table with relative frequencies.
- A joint relative frequency is the proportion of individuals who share two characteristics at once.
- A conditional relative frequency is a proportion within one row or one column.
If the conditional proportions differ from group to group, the variables appear associated.
Graphical Representations
Side-by-Side Bar Graphs
Side-by-side bar graphs place the bars for each category of one variable next to each other, grouped by the categories of the other variable. By comparing bars within each group, you can see how the categories of one variable line up against the other.
Segmented Bar Graphs
Segmented bar graphs show each group as one full bar split into category pieces. Since each bar adds to 100%, they are especially useful when you want to compare conditional distributions.
Mosaic Plots
Mosaic plots divide the display into rectangles whose areas are proportional to the data. Width and height both carry information, so the plot functions like a visual version of a two-way table.
Deciding Whether Variables Appear Associated
To decide whether two categorical variables appear associated, compare the conditional proportions across groups. If the distributions look noticeably different, the variables appear associated. If the conditional distributions are very similar, the variables may be approximately independent.
For example, suppose you have data on class level and whether students finish homework on time. If the proportion of on-time students changes noticeably across grade levels, that suggests an association between the variables.
Keep in mind that an apparent pattern may still be due to chance, and even a real association does not prove causation.
How to Use This on the AP Statistics Exam
MCQ
- Read carefully to tell whether the question wants a joint relative frequency or a conditional one.
- Practice pulling cells, row totals, and column totals from a two-way table quickly.
- When the question asks whether variables are associated, compare proportions, not totals.
Free Response
- State what the table or graph shows in context, using the actual variable names.
- To argue for or against association, compare conditional proportions instead of raw counts.
- Use cautious language such as "appears associated" rather than causal language.
Common Trap
Comparing raw counts instead of proportions when groups have different sizes. Association questions almost always need conditional distributions.
Common Misconceptions
- A joint relative frequency is not the same as a conditional one. Dividing by the grand total and dividing by a row or column total answer different questions.
- Equal counts mean no association. Groups often have different sizes, so counts alone can mislead.
- An association proves causation. It does not.
- A mosaic plot is just a fancy bar chart. Widths matter too.
Related AP Statistics Guides
Vocabulary
The following words are mentioned explicitly in the AP® course framework for this topic.Term | Definition |
|---|---|
association | The relationship between two variables where knowing the value of one variable provides information about the other variable. |
distribution | The pattern of how data values are spread or arranged across a range. |
joint relative frequency | A cell frequency in a two-way table divided by the total number of observations in the entire table, expressing the proportion of the total for a specific combination of categories. |
mosaic plots | A graphical representation of two categorical variables where rectangles are sized proportionally to represent the frequency or relative frequency of each combination of categories. |
segmented bar graphs | A graphical representation where bars are divided into segments, with each segment representing a category of a second categorical variable, showing the composition within each category of the first variable. |
side-by-side bar graphs | A graphical representation that displays bars for one categorical variable grouped side-by-side for each category of another categorical variable, allowing for easy comparison between groups. |
two-way table | A table that displays the frequency distribution of two categorical variables, organized in rows and columns. |
Frequently Asked Questions
How do you represent two categorical variables in AP Stats?
Use a two-way table, also called a contingency table, or a graph such as a side-by-side bar graph, segmented bar graph, or mosaic plot. These displays show how one categorical variable breaks down across another.
What is a two-way table?
A two-way table summarizes two categorical variables using rows and columns. Each cell contains a frequency count or relative frequency for one combination of categories.
What is joint relative frequency?
A joint relative frequency is one cell count divided by the total number of individuals in the table. It describes the proportion of the whole sample that falls into a specific combination of categories.
What is the difference between marginal and conditional relative frequency?
A marginal relative frequency uses a row total or column total out of the grand total. A conditional relative frequency looks within one row or column category, so the denominator is that row or column total.
When should you use a segmented bar graph?
Use a segmented bar graph when you want to compare conditional distributions across groups. Each bar is scaled to the same total, which makes the proportions within categories easier to compare.
What is a common AP Stats mistake with two categorical variables?
A common mistake is comparing raw counts when group sizes are different. Use relative frequencies or conditional distributions when you need a fair comparison across groups.