top of page

Proposed Changes to the AP Statistics Course Framework

Updated: May 12

The College Board released the first ever Course and Exam Description (CED) for AP Statistics in 2019, which provided clear specificity in what needed to be covered in the course and what would be assessed on the AP Exam. Here we are five years later, and it looks like we are going to get some updates. Just today, the College Board proposed a draft of the revised course framework, to be implemented for the 2026-2027 school year and assessed for the first time on the 2027 AP Statistics Exam. The College Board is asking us AP Stats teachers to review the proposed changes and provide feedback before May 27,2024.


What Are the Proposed Big Picture Changes?


Fewer Units and a Different Order

The current CED has 9 Units. The new proposed Course Framework is condensed to only 5 units. The guiding philosophy here is that reducing the amount of content that needs to be covered will allow teachers to go more in-depth into each topic. Here is a quick correlation guide.


Course Skills are Being Replaced by Statistical Practices

In the current CED, there are four course skills categories (Selecting Statistical Methods, Data Analysis, Using Probability and Simulation, and Statistical Argumentation). In the new proposed course framework, there are four statistical practices (Formulate Questions, Collect Data, Analyze Data, Interpret Results). The four statistical practices come directly from the Guidelines for Assessment and Instruction in Statistics Education (GAISE), a framework for statistics education proposed by the American Statistical Association.


What Content Has Been Removed?

Our first glance through the new proposed course framework shows the "less is more" approach. Here are the topics that have been removed.

  • Cumulative frequency graphs

  • Outliers, high-leverage points, and influential points

  • Analyzing departures from linearity

  • Geometric distributions

  • Estimating probabilities using simulations

  • Transforming and combining random variables

  • Type I, Type II Errors, and Power

  • Chi-square goodness-of-fit test

  • Inference for slope

What Content Has Been Added?

There is not a lot to report here, but we did find one Learning Objective that seems to be a new one:

  • Determine a research question within a statistical study. (1.1.B)


Where Do I Provide Feedback?

The College Board is looking for teachers to review the new proposed course framework and then to provide detailed feedback for review. You can submit your feedback on the proposed course framework here.


Will Stats Medic Make Adjustments?

Of course! You know we got your backs. We will stay on top of all the changes as they happen, and you can be sure that we will have perfectly aligned lessons along with homework and assessments ready to go the summer before the 2026-2027 school year!

6,298 views10 comments

Recent Posts

See All

10 bình luận


Robert Aronstam
Robert Aronstam
01 thg 5

I'd like to know if the StatsMedic's "O Holistic One" predicts that these removed topics will be de-emphasized on the AP Exam during the next two school years (before the official adoption of the new framework).

Thích
Thành viên không xác định
01 thg 5
Phản hồi lại

Because the AP Exams take several years to create, my guess is that the next two exams are already well into production and I would expect them to cover all the content that exists in the current CED.

Thích

Sebastian Pugliese III
Sebastian Pugliese III
01 thg 5

In mathematics, less is not more. Less is less.

Thích

je175
je175
01 thg 5

In my reading of the draft, simulating to find a P-value is still present:

3.6.A.2 "If the distribution of the test statistic has been simulated, the P-value..."


See also, 2.8.A.3 on simulating a probability distribution... and we still get to simulate sampling distributions (2.12.A.2)


I'm a little sad that standard deviation of the residuals is on the chopping block, but I get it. It becomes "just another interpretation" to memorize. It has been nice as an application of standard deviation and thinking about residuals.

Thích

Amy Hogan
Amy Hogan
01 thg 5

Thanks for the summary of what's been removed.


I like the order pedagogically (and this is how I have been teaching it for many years now). It makes sense to have all proportions together, means together and then that flows nicely into regression with two variables. Not sure why fewer units... I think that makes it harder for teachers who like giving unit tests or having a specific set of textbook chapters to assess. I can parse however I want so that doesn't bother me as much.


I am sort of scratching my head about the removal of errors/power, random variable transformations, and the influence of extreme points in two-variable data. But, most of all, not sure why simulations are…


Thích
Thành viên không xác định
01 thg 5
Phản hồi lại

I couldn't agree with you more about being concerned with Error types being removed. It is imperative for students to understand the relationship between significance level and probability of Type I error. Otherwise they just think lower sig level would always be better. It does not take a lot of time to cover this topic as a part of any inference but it enhances understanding of statistical results and potential issues (a chance to reinforce the consequences of variability in Sampling and the fact that inferences can be incorrect.

I feel the same way about interpreting the results of simulations , they create a tangible method to experience the meaning of p-value which enhances their understanding of this potentially confusing…

Thích

Thành viên không xác định
01 thg 5

Let's call it what it is: they're watering down the course.

Thích
bottom of page