Teaching statistics becomes more concrete when instructors connect curriculum design with assessment structure and observed student outcomes. AP Statistics provides a defined nine-unit sequence, a balanced exam, recent free-response scoring data, and long-term participation measures that can help teachers organize instruction and discuss scale.
Contents
- Course structure and unit weights
- Exam format and timing
- What the 2025 free-response data shows
- The 2024 score distribution
- How AP participation has changed
- Using the figures in teaching plans
Course structure and unit weights
The AP Statistics Course describes nine commonly taught units of study. The published multiple-choice section weights indicate how much emphasis each unit receives on that section of the exam. These are ranges rather than single fixed percentages, so they are best treated as planning bands.
| Unit | Topic | Multiple-choice weight |
|---|---|---|
| 1 | Exploring One-Variable Data | 15%–23% |
| 2 | Exploring Two-Variable Data | 5%–7% |
| 3 | Collecting Data | 12%–15% |
| 4 | Probability, Random Variables, and Probability Distributions | 10%–20% |
| 5 | Sampling Distributions | 7%–12% |
| 6 | Inference for Categorical Data: Proportions | 12%–15% |
| 7 | Inference for Quantitative Data: Means | 10%–18% |
| 8 | Inference for Categorical Data: Chi-Square | 2%–5% |
| 9 | Inference for Quantitative Data: Slopes | 2%–5% |
The largest upper-end range is Unit 1, Exploring One-Variable Data, at 23%. Unit 4, Probability, Random Variables, and Probability Distributions, reaches 20%, while Unit 7, Inference for Quantitative Data: Means, reaches 18%. These ranges suggest that exploratory data analysis, probability, and inference for means deserve sustained attention in a teaching sequence. The AP Statistics Course is the source for all nine unit labels and weights.
At the lower end, Unit 8, Inference for Categorical Data: Chi-Square, and Unit 9, Inference for Quantitative Data: Slopes, each have a 2%–5% multiple-choice range. Their lower published ranges do not make them optional topics: both are part of the nine-unit course structure, and both can require students to connect vocabulary, conditions, calculations, and interpretation.
AP Statistics revisions launch in the 2026–27 school year, according to the AP Statistics Course. Teachers using materials across school years should therefore keep the revision timing explicit when comparing course plans or assessment resources.
Exam format and timing
The AP Statistics Exam has two equally weighted sections. The multiple-choice section contains 40 questions, is worth 50% of the score, and lasts 1 hour 30 minutes. The free-response section contains six questions, is worth the other 50%, and also lasts 1 hour 30 minutes. These figures come from the AP Statistics Exam description.
The free-response section is divided into two parts. Part A contains five multipart questions with specific focuses on collecting data, exploring data, probability and sampling distributions, inference, and a mixed-skill item. Part B contains one investigative task. Each of the six 2025 free-response questions was worth four possible points, according to AP Statistics Scoring Statistics 2025.
The equal time allocation creates a useful teaching frame: students need both selected-response fluency and written statistical communication. A class plan that concentrates only on procedures may leave students underprepared for multipart explanations and the investigative task. Conversely, written practice can be organized around the same broad content areas named for Part A: collecting data, exploring data, probability and sampling distributions, inference, and mixed skills.
The AP Statistics Exam was scheduled for Thursday, May 7, 2026, at 12 PM local time. That is a scheduled exam date and time, not a general claim about every administration. The date is supplied by the AP Statistics Exam source and should be kept separate from the broader, recurring exam design.
What the 2025 free-response data shows
AP Statistics Scoring Statistics 2025 reports a mean score and standard deviation for each question. Every question used a four-point scale. The figures below describe performance on those specific 2025 questions; they do not establish a timeless ranking of student ability or a forecast for another administration.
| Question | Mean score, out of 4 | Standard deviation |
|---|---|---|
| FRQ 1 | 1.75 | 1.13 |
| FRQ 2 | 2.10 | 1.18 |
| FRQ 3 | 1.55 | 1.18 |
| FRQ 4 | 1.65 | 1.44 |
| FRQ 5 | 1.43 | 1.21 |
| FRQ 6 | 1.78 | 1.21 |
FRQ 2 had the highest reported mean at 2.10 out of 4. FRQ 5 had the lowest at 1.43, followed by FRQ 3 at 1.55 and FRQ 4 at 1.65. FRQ 1 had a mean of 1.75, while FRQ 6 had a mean of 1.78. Since each item had four possible points, the reported means show that the average score was below half of the available points on all six questions.
The standard deviations also varied. FRQ 4 had the largest reported standard deviation, 1.44. FRQ 1 had a standard deviation of 1.13; FRQ 2 and FRQ 3 each had 1.18; and FRQ 5 and FRQ 6 each had 1.21. For classroom discussion, that spread is a reminder that a question mean alone does not describe how consistently students performed. The 2025 scoring figures include both the central result and the variation measure.
These data can support targeted review conversations. The lowest mean in this particular set was FRQ 5, but the source does not identify a universal cause for that result. Teachers can use the item focus and student work from their own classes to investigate whether difficulties involve setup, conditions, calculation, interpretation, or communication. The published figures should be presented as 2025 scoring statistics, not as independently verified legacy facts or as causal evidence.
The 2024 score distribution
AP Score Distributions 2024 reports results for 177,819 students taking the 2024 AP Statistics exam. A total of 147,498 students scored 3 or higher, representing 82.9% of students in that reported distribution. The mean score was 3.54 and the standard deviation was 1.07.
The distribution included 37,662 students earning a 5 and 55,833 earning a 4. Another 54,003 students earned a 3, 24,900 earned a 2, and 5,421 earned a 1. The reported percentages include 21.2% earning a 5, 22.5% earning a 3, 15.9% earning a 2, and 11.4% earning a 1. The supplied 2024 distribution figures do not include a percentage for the 4 category in the facts used here, so no additional percentage is inferred.
| 2024 score | Students |
|---|---|
| 5 | 37,662 |
| 4 | 55,833 |
| 3 | 54,003 |
| 2 | 24,900 |
| 1 | 5,421 |
| Total | 177,819 |
For teaching statistics, the 2024 results offer several distinct reference points: the full tested population, the count at or above 3, the mean, and the standard deviation. They should not be treated as a direct measure of an individual classroom, because the figures describe the 2024 AP Statistics exam population. They also should not be used to infer why students received particular scores without additional evidence.
How AP participation has changed
Annual AP Program Participation 1956–2024 reports participation at selected school-year points. In 1955–56, AP included 104 schools, 1,229 students, 2,199 exams, and 130 colleges. By 1965–66, the figures were 2,518 schools, 38,178 students, 50,104 exams, and 1,076 colleges. In 1975–76, AP had 3,937 schools, 75,651 students, 98,898 exams, and 1,580 colleges.
The selected milestones continued upward in later decades. In 1985–86, AP had 7,201 schools, 231,378 students, 319,224 exams, and 2,125 colleges. In 1995–96, it had 11,712 schools, 537,428 students, 843,423 exams, and 2,895 colleges. In 2005–06, the reported totals were 16,000 schools, 1,339,282 students, 2,312,611 exams, and 3,638 colleges.
| School year | Schools | Students | Exams | Colleges |
|---|---|---|---|---|
| 2010–11 | 18,340 | 1,973,545 | 3,456,020 | 4,001 |
| 2014–15 | 21,594 | 2,483,452 | 4,478,936 | 4,154 |
| 2018–19 | 22,678 | 2,825,710 | 5,098,815 | 4,361 |
| 2020–21 | 22,802 | 2,548,228 | 4,578,302 | 3,024 |
| 2023–24 | 23,722 | 3,079,134 | 5,744,259 | 4,186 |
The same source reports 2019–20 totals of 22,152 schools, 2,642,630 students, 4,751,957 exams, and 3,160 colleges. In 2021–22, the totals were 22,977 schools, 2,659,914 students, 4,762,347 exams, and 3,820 colleges. In 2022–23, AP had 23,071 schools, 2,869,418 students, 5,197,601 exams, and 3,907 colleges. The 2023–24 figures were 23,722 schools, 3,079,134 students, 5,744,259 exams, and 4,186 colleges.
Across the period from 1955–56 to 2023–24, Annual AP Program Participation 1956–2024 reports cumulative participation of 57,707,612 students and 100,209,322 exams. These are cumulative program figures, so they measure participation volume across the stated period rather than a single-year enrollment count.
Using the figures in teaching plans
The course weights can guide calendar conversations, while the exam structure can guide practice formats. Unit ranges identify the relative assessment emphasis reported by the AP Statistics Course; they do not prescribe the number of class days for any unit. The two 90-minute exam sections make it reasonable to include both timed selected-response work and timed written responses in preparation, while keeping the distinction between classroom practice and official exam conditions clear.
The 2025 free-response means and standard deviations can serve as an assessment-literacy example. Students can read a mean score alongside a standard deviation, identify what each describes, and discuss why neither statistic alone explains performance. The 2024 score distribution can support a separate lesson on counts, percentages, mean, and standard deviation, provided the geography and measurement context remain explicit: these are results for students taking the 2024 AP Statistics exam.
Finally, the annual participation figures provide historical context for teaching statistics at program scale. They show reported counts of schools, students, exams, and colleges at selected school years, with a cumulative total through 2023–24. Keeping school year, exam year, and cumulative period visible helps students distinguish a point-in-time measurement from a long-run total.