A categorical variable is summarized by how many individuals fall in each category (the frequency) and what fraction of the total that is (the relative frequency). Edit the table, watch every proportion update, and practice the backward problems AP likes to ask.
A frequency table lists each category with its count. Divide each count by the total to get the relative frequency.
Type in the count column. Everything else updates.
AP questions often give percents and ask for counts, or leave one cell blank. Solve, then check.
Lists each category and how many individuals are in it. The counts add to the total number of individuals, n.
Count ÷ total. It can be written as a decimal (0.36) or percent (36%). The relative frequencies of all the categories add to 1, apart from rounding.
Proportions put groups of different sizes on the same scale. 60 out of 100 and 60 out of 400 are very different results.
The category with the highest frequency. It is the only measure of center that makes sense for categorical data.
Count = proportion × total, and total = count ÷ proportion. Many AP questions give you two of the three.
“36% of the 250 surveyed students chose fall” is a complete summary: statistic, group, and size.
Write the formula, plug in, then answer in context.
| Question | Formula | Work | Answer |
|---|---|---|---|
| 48 of 120 students ride the bus. Relative frequency? | count ÷ total | 48 ÷ 120 | 0.40 (40% of the students) |
| 35% of 600 shoppers paid with cash. How many? | proportion × total | 0.35 × 600 | 210 shoppers |
| 72 people, which is 18% of the sample, chose Rock. Sample size? | count ÷ proportion | 72 ÷ 0.18 | 400 people |
| Relative frequencies 0.5, 0.2, 0.2, and ? | 1 − sum of the others | 1 − 0.9 | 0.1 |
AP-style multiple choice. Pick an answer to see the explanation.
The vocabulary for topic 1.3. Read the list, flip through flashcards, or quiz yourself. For every term in the unit, see the Unit 1 terminology guide.