Quadratic regression is a technique that uses technology to fit a quadratic (degree-2 polynomial) function to a data set, chosen when the data shows one change of direction or a linearly changing rate of change, as covered in AP Precalculus Topic 1.14.
Quadratic regression is what you reach for when data curves like a parabola. Instead of guessing coefficients by hand, you enter the data into your calculator, run a quadratic regression, and the technology finds the best-fit function of the form f(x) = ax² + bx + c.
The real skill the CED cares about isn't pushing the button. It's recognizing when a quadratic model is the right choice. Per the essential knowledge for 1.14.A, the choice of function type should be based on the pattern in the data and the context of the problem. Quadratic data has a signature look. The output values rise then fall (or fall then rise) exactly once, and the rate of change itself changes at a constant rate. A classic giveaway is when the second differences of evenly spaced data are constant. Think projectile height over time. It goes up, peaks, and comes back down. One turn, one parabola, quadratic regression.
Quadratic regression lives in Topic 1.14 (Function Model Construction and Application) in Unit 1, and it directly supports learning objectives 1.14.A (construct a polynomial function model), 1.14.B (use a function model, where the essential knowledge explicitly names quadratic regression as a tool), and 1.14.D (apply the model to predict values and rates of change). Topic 1.14 is the payoff of Unit 1. Everything you learned about polynomial behavior, end behavior, and rates of change gets cashed in here to pick the right model for real data. Quadratic regression is also where AP Precalculus tests your calculator skills in context, since regression is one of the technology tasks the course expects you to perform fluently. For the full modeling picture, the Topic 1.14 study guide covers all the model types side by side.
Keep studying AP® Precalculus Unit 1
Regression analysis (Unit 1)
Quadratic regression is one specific case of regression analysis, the general process of fitting a function to data. The CED lists linear, quadratic, cubic, and quartic regressions as the family you choose from, and the degree you pick should match the pattern in the data.
Quadratic function (Unit 1)
Quadratic regression outputs a quadratic function, so everything you know about parabolas transfers. The model has exactly one extremum, its rate of change is linear, and its second differences are constant. Those properties are exactly what tell you a quadratic fit is appropriate in the first place.
Cubic regression (Unit 1)
Cubic regression is the next step up the polynomial ladder. If your data changes direction twice instead of once, a quadratic can't capture it and you need degree 3. Counting the turns in the data is the fastest way to decide between them.
Piecewise-defined function (Unit 1)
Per 1.14.B, you can combine modeling techniques into a piecewise-defined model. A scenario might be quadratic on one interval (a ball in flight) and constant or linear on another (after it lands), so quadratic regression can build just one piece of a bigger model.
Quadratic regression shows up most often in model-selection questions. A typical multiple-choice stem gives you a data table or describes a residual plot and asks which regression model fits best. Two patterns to memorize. First, if a linear regression leaves a residual plot with a clear U-shape, that's the signal to try quadratic regression instead, because the leftover pattern is itself parabolic. Second, if a data set's second differences are roughly constant, or the context involves one peak or one valley, quadratic is the answer. You should also be ready to use a quadratic model once it's built (LO 1.14.D), meaning predict a value, find an average rate of change over an interval, and attach the correct units pulled from the context. No released FRQ has used the phrase verbatim, but Topic 1.14 modeling skills feed directly into the calculator-active portions of the exam.
Linear regression fits data with a constant rate of change, so the residuals should look like random scatter. Quadratic regression fits data whose rate of change is itself changing at a constant rate. The exam loves this distinction. If you fit a line and the residual plot bends into a U-shape, the line missed the curvature, and quadratic regression is the appropriate next model. Constant first differences mean linear; constant second differences mean quadratic.
Quadratic regression uses technology to fit a degree-2 polynomial model f(x) = ax² + bx + c to a data set.
Choose quadratic regression when the data changes direction exactly once or when the second differences of evenly spaced data are roughly constant.
A U-shaped pattern in the residual plot of a linear regression is the classic signal that you should try a quadratic model instead.
The CED (1.14.B) explicitly lists quadratic regression alongside linear, cubic, and quartic regression as technology-based modeling tools.
Once you have the model, LO 1.14.D expects you to use it to predict values and rates of change, with units pulled from the context.
Data with two changes of direction needs cubic regression, not quadratic, so count the turns before you pick a degree.
It's a technique where technology fits a quadratic function (ax² + bx + c) to a data set. It appears in Topic 1.14 under learning objectives 1.14.A and 1.14.B as one of the standard polynomial regression types.
Use quadratic regression when the data curves with one change of direction, when second differences are roughly constant, or when a linear fit leaves a U-shaped residual plot. Linear regression only works when the rate of change is constant.
Not exactly. A quadratic function is the mathematical object itself, while quadratic regression is the process of using technology to find the specific quadratic function that best fits a set of data points.
No. The CED says regression models are constructed using technology, so your calculator does the fitting. The exam tests whether you can choose the right model type and then interpret and apply the result.
Quadratic regression fits a degree-2 model for data with one change of direction, while cubic regression fits a degree-3 model for data with two changes of direction. Counting how many times the data turns tells you which degree to use.
Connect this key term to the AP exam workflow: review the course, practice questions, and check related study tools.
Review units, study guides, and course resources.
Check this vocabulary in multiple-choice context.
Apply key concepts in written AP responses.
Estimate the exam score you are working toward.
Review the highest-yield facts before practice.
Put the full course together before test day.