- Unit weight: the smallest unit on the official exams, 27 of the 406 multiple-choice questions on today’s course (6.7%) and about 94 of the 1,040 free-response points we counted (9.1%), but the 2026 CED gives it 10–20% of the multiple choice.
- Top types: computing or interpreting a residual and what correlation means (7 official questions each), then interpreting the slope or intercept; on the free response, describing a scatterplot.
- Practice: Unit 5 is 10.4% of the practice questions we tagged but 6.7% of the official ones. Properties of the least-squares line are over-drilled about 6 times over, while residuals and interpreting the slope get less than their share.
Unit 5 is the smallest unit of the AP Statistics multiple choice so far, and the course description expects more of it. College Board’s Course and Exam Description (the CED) gives it 10–20% of the multiple choice. On the six official practice exams from 2012–2019 it was 10 of 215 questions (4.7%), 0 to 2 per exam. In the 24 sample questions of the 2026 CED, the only official questions in the new format, it is 3.
The unit is descriptive regression: scatterplots, correlation, the least-squares line, residuals and . It used to be part of the old Unit 2. Inference for the slope of a regression line is no longer in the course, and neither are influential points or transformations to straighten a curved pattern.
The exam changes in May 2027. Section I becomes 42 multiple-choice questions with four choices instead of five, and some come in sets: two or three questions on one shared stimulus (6 of the 24 CED sample questions are in 2 sets). A scatterplot with its least-squares line or computer output, then questions on the slope, a prediction and , is a natural set. Section II becomes 4 free-response questions of 10 points each, scored point by point. At the 2012–2019 rate, about 2 of the 42 multiple-choice questions would be Unit 5; the CED’s 10–20% range puts it at about 4 to 8.
On the free response, Unit 5 earned about 94 of the 1,040 points on the 124 official questions we counted, and 15 of those 124 questions had a Unit 5 part. None of the 40 points in the four 2026 sample questions is on Unit 5.
Unit 5 on the free response: Question 2
From 2027 the free-response questions come in a fixed order: Question 1 assesses Practices 1 and 2 (the investigative question and how data are collected), Question 2 Practices 3 and 4 (analyzing data and drawing conclusions), Question 3 inference, and Question 4 a mix of Practices 2, 3 and 4.
Regression lives in Question 2. Of the 54 official free-response questions that fit the new Question 2, 12 had a Unit 5 part, most recently on the 2026 exam; of the 22 that fit Question 4, none did. A regression was the whole setting of 7 operational free-response questions from 2012 to 2026 (2012, 2015, 2016, 2017, 2018, 2022 and 2023): a scatterplot to describe, a least-squares line or computer output to read, and a slope, a prediction or a residual to interpret in context.
The Unit 5 question types the multiple choice asks
Ranked by the number of official multiple-choice questions on today’s course (13 College Board sources, 1997–2026; ties by the number of sources, then the practice exams), with how many of the 13 sources had one, how many were on the six 2012–2019 practice exams, and how many of the 84 operational free-response questions since 2012 had a part on it. Seven types have an official multiple-choice question.
1. Computing or interpreting a residual
A residual is observed minus predicted, : positive when the point is above the line, negative when it is below.
Show solutionHide solution
(A) . The line predicts minutes for a route with 28 stops. The residual is observed minus predicted: minutes, so this route is scheduled 12.02 minutes faster than the line predicts.
Practice: Line of Best Fit worksheet · AP Statistics Unit 5 worksheet
2. What correlation means
The correlation measures the direction and strength of a linear association. It has no units, is between and , and does not change when the units change or the variables switch roles.
- There is a strong, positive linear association between trail length and hiking time.
- About 91.6% of the hikes lie on the least-squares regression line.
- About 91.6% of the variation in hiking time is explained by the linear relationship with trail length.
- For each additional kilometer of trail, hiking time increases by 0.916 minutes.
Show solutionHide solution
(A) There is a strong, positive linear association between trail length and hiking time. The sign of gives the direction, positive, and its closeness to 1 the strength: a strong, positive linear association. Longer hikes tend to take longer.
Practice: Correlation Coefficient worksheet · AP Statistics Unit 5 worksheet
3. Interpreting the slope or the intercept in context
The slope is the predicted change in the response for each one-unit increase in the explanatory variable; the intercept is the predicted response when the explanatory variable is 0, which means something only when 0 is within the data.
- For each additional percentage point of screen brightness, the predicted battery life decreases by 0.0529 hours.
- For each one-unit increase in battery life, the predicted screen brightness decreases by 0.0529 percent.
- The predicted battery life is hours when the screen brightness is 0.
- For each additional percentage point of screen brightness, the battery life decreases by exactly 0.0529 hours.
Show solutionHide solution
(A) For each additional percentage point of screen brightness, the predicted battery life decreases by 0.0529 hours. The slope is the coefficient of Brightness, , and it describes the line, so the change is a predicted one: 0.0529 hours, about 3 minutes, less battery life for each point of brightness.
Practice: Line of Best Fit worksheet · AP Statistics Unit 5 worksheet
4. Describing a scatterplot
Give the direction, form and strength of the association and any unusual features, in context.
- There is a strong, positive, linear association between amount of nitrogen fertilizer and corn yield, with no unusual points.
- There is a strong, negative, non-linear (curved) association between amount of nitrogen fertilizer and corn yield, with no unusual points.
- There is a strong, positive, non-linear (curved) association between amount of nitrogen fertilizer and corn yield, with no unusual points.
- There is a weak, positive, non-linear (curved) association between amount of nitrogen fertilizer and corn yield, with no unusual points.
Show solutionHide solution
(C) There is a strong, positive, non-linear (curved) association between amount of nitrogen fertilizer and corn yield, with no unusual points. Plots that received more nitrogen tend to have higher yields, so the association is positive. The points follow a curve that rises quickly and then levels off, so the form is non-linear, and they stay close to that curve, so it is strong. No point stands apart from the pattern.
Practice: Scatter Plots worksheet · AP Statistics Unit 5 worksheet
5. Interpreting
, R-Sq in computer output, is the percent of the variation in the response that is explained by the linear relationship with the explanatory variable.
- About 88.4% of the variation in trail length is explained by the linear relationship with hiking time.
- About 88.4% of the hikes lie on the least-squares regression line.
- About 88.4% of the variation in hiking time is explained by the linear relationship with trail length.
- The correlation between trail length and hiking time is 0.884.
Show solutionHide solution
(C) About 88.4% of the variation in hiking time is explained by the linear relationship with trail length. R-Sq is : about 88.4% of the variation in hiking time is explained by the linear relationship with trail length. The rest of the variation comes from other things, such as how steep the trail is.
Practice: Correlation Coefficient worksheet · AP Statistics Unit 5 worksheet
6. Using a residual plot to judge a linear model
A residual plot with random scatter around 0 supports a linear model; a curve in the residuals means a curved relationship, however close is to 1 or .
- Yes. The residuals are both positive and negative, so a linear model is appropriate.
- Yes. The correlation is close to , so a linear model is appropriate.
- No. The residual plot shows a curved pattern, so a linear model is not appropriate.
- No. The residual plot shows a curved pattern, so there is no association between screen brightness and battery life.
Show solutionHide solution
(C) No. The residual plot shows a curved pattern, so a linear model is not appropriate. The residuals are above 0 at the lowest and highest brightness levels and below 0 in the middle: a curved pattern. The relationship between brightness and battery life bends, so a linear model is not appropriate, even with .
Practice: Line of Best Fit worksheet · AP Statistics Unit 5 worksheet
7. Properties of the least-squares line
The least-squares line makes the sum of the squared residuals as small as possible and passes through , so and .
Show solutionHide solution
(A) . The line passes through , so minutes.
Practice: Line of Best Fit worksheet · AP Statistics Unit 5 worksheet
On the free response: what multiple choice rarely asks
- Describing a scatterplot. 3 official multiple-choice questions, but 11 operational free-response questions from 2012 to 2026 had a part on it, 4 of them since 2019 and most recently in 2026. It is the most common Unit 5 part on the free response.
- Predicting with the line. No official multiple-choice question, but 3 operational free-response questions had a part on it, 2 of them since 2019 (most recently in 2026). Substitute the value into the equation, give the units, and say whether the value is inside the data’s range.
- Interpreting the slope in context. 5 official multiple-choice questions and 6 operational free-response questions, 3 of them since 2019. The words that earn the point: “predicted” (or “on average”), the value with its units, and both variables in context.
Show solutionHide solution
(D) . From the output, . For 111 kilograms of nitrogen per hectare, tonnes per hectare. 111 is inside the data’s range of nitrogen amounts, so this is an interpolation.
Rising and fading in Unit 5
We compare the 42 operational free-response questions of 2012–2018 with the 42 of 2019–2026 (both 7 exams; 2020 was never released).
- Unit 5 as a whole is fading on the free response: 15.1% of the free-response points in 2012–2018, 7.3% in 2019–2026 (7 questions with a Unit 5 part, then 4).
- Fading: describing a scatterplot went from 7 questions to 4, though it is still the most common Unit 5 part.
- Steady: the slope, residuals and . Interpreting the slope or intercept was in 3 questions in each period, residuals in 2 and in 1.
- Rising slightly: predicting with the line, from 1 question to 2.
The course description pulls the other way: its 10–20% for Unit 5 is above the 6.7% of the official multiple choice, and the new Question 2 is where a full regression question fits.
Not asked yet: interpolation and extrapolation
Topic 5.3 names interpolation (predicting inside the range of the data) and extrapolation (predicting outside it) for the first time. No official multiple-choice question has asked about them; one free-response part on a 2018 practice exam asked why an extrapolation might not be appropriate.
Dropped from the course: inference for the slope
The interval and test for the slope of a regression line are gone from the 2026 course. 4 of the official free-response questions we counted had a part on them: 1 on an exam (2023) and 3 on practice exams. Old review books still teach them; you can skip those pages.
What students over- and under-practice in Unit 5
We tagged 1,130 multiple-choice questions from teacher tests, review packets, AP Classroom exports and prep books (reposted official questions removed) and compared each type’s share of that practice with its share of the 406 official questions. Unit 5 is 10.4% of the practice questions against 6.7% of the official ones.
| Question type | Practice vs official share | Ratio |
|---|---|---|
| Properties of the least-squares line | 1.59% vs 0.25% | 6.5× over |
| Use a residual plot to judge a linear model | 0.71% vs 0.25% | 2.9× over |
| What correlation means | 3.27% vs 1.72% | 1.9× over |
| Describe a scatterplot | 1.24% vs 0.74% | 1.7× over |
| Interpret | 0.62% vs 0.74% | 0.8× under |
| Compute or interpret a residual | 1.24% vs 1.72% | 0.7× under |
| Interpret the slope or intercept | 0.80% vs 1.23% | 0.6× under |
Official counts per type are small (1 to 7 questions), so read the ratios as directions.
- Least-squares mechanics are over-drilled. Properties of the line (the slope from , and , the point of means) have a single official multiple-choice question but get six and a half times that share of practice.
- Residuals and the slope in context are under-practiced. Together they have 12 official multiple-choice questions, and the slope is in 6 operational free-response questions since 2012, but each gets less than its share.
- Correlation is practiced about twice as much as it is asked, which is fine: the multiple choice asks it as often as residuals.
How we counted
Every number here comes from public official AP Statistics material: 406 multiple-choice questions on today’s course from 13 College Board sources (the 1997, 2002 and 2007 released exams, the 2008, 2012–2015, 2018 and 2019 practice exams, and the sample questions in the 2005, 2020 and 2026 course descriptions), and 124 free-response questions (the 2012–2026 exams except 2020, the practice exams and the four 2026 samples; 1,040 points). Free-response points before 2027 are split from each question’s holistic score in proportion to its scoring sections. Each question is tagged to its CED topic and question type. Every example problem is our own.
Practice Unit 5
Our free AP Statistics Unit 5 practice test has 20 original multiple-choice questions with four choices, including two sets of questions on one shared stimulus (a least-squares line, then its slope and a prediction; a scatterplot, then its description), weighted by how often the official exams ask each type. It comes with an answer key, worked solutions and a printable PDF.
For free response, the AP Statistics Unit 5 FRQ set has 2 questions: one in the style of Question 2 of the 2027 exam, where every official regression free-response question fits (a regression read from computer output, a scatterplot and a residual plot), and one in the style of Question 4, a regression study with questions on its sample and its scope. Each is scored out of 10 points, with a scoring guide and a PDF. Both make good practice before the Unit 5 Progress Check in AP Classroom.
The formulas this unit uses, each with a worked example and the calculator command, are on our free AP Statistics formula sheet.
The previous unit: the AP Statistics Unit 4 review (inference for means) and the Unit 4 worksheet. For the whole course in one place, see the AP Statistics review.