An influential point is a data point that changes a statistical result a lot, especially in regression. In Intro to Statistics, you look for it because one point can pull the line, shift predictions, and distort the story your data seems to tell.
An influential point is a data value in Intro to Statistics that has a much bigger effect on a model than the rest of the points do. The clearest place you see it is in regression, where one observation can noticeably change the slope, the intercept, or how well the line fits the data.
Not every unusual point is influential. A point can be far away from the pattern and still not change the regression line much if it sits near the center of the x values. What makes a point influential is its effect on the model, not just how strange it looks. That is why students often compare outliers and influential points, because they are related but not identical.
A big reason influential points matter is leverage. If a point has an extreme x value, it has high leverage, which means it has more pull on the line than a point near the middle of the data. A high-leverage point can bend the regression line toward itself, which changes the predicted y values for lots of other x values too.
You can see this in a simple scatterplot. Imagine most of the points follow an upward trend, but one point sits far to the right and much higher than expected. If you include that point, the line may tilt upward more steeply than it would otherwise. If you remove it, the slope might flatten and the predictions change a lot. That kind of change is what makes the point influential.
Intro Stats courses often talk about influential points alongside residuals and diagnostic checks. A large residual tells you a point is not fitting the model well, but an influential point can also have a small residual and still strongly affect the line if it has high leverage. That is the part students miss most often. The goal is not to delete every unusual point, but to notice when one observation is steering the analysis more than it should.
Influential points matter in Intro to Statistics because regression is only useful when the model reflects the overall pattern in the data. If one point has too much control, your slope, intercept, correlation, and predictions can all shift in misleading ways.
This shows up any time you interpret a scatterplot or run a line of best fit. A class problem might ask whether a new data point changes the regression equation, whether a prediction still makes sense, or whether a relationship looks stronger just because of one extreme observation. If you ignore the influential point, you may describe the trend incorrectly.
It also matters for outlier work. Many students think every outlier should be removed, but that is not how statistics works. Sometimes a point is unusual because it is real and meaningful, like a rare case in a dataset. Other times it is a recording error or a bad measurement. Influential points force you to ask which situation you have and whether the model should be reported with or without that point.
This idea also connects to good statistical communication. When you write about a regression result, you need to mention if the line depends heavily on one observation. That makes your conclusion more honest and helps you explain the limits of the model instead of treating it like a perfect summary of the data.
Keep studying Intro to Statistics Unit 12
Visual cheatsheet
view galleryOutlier
An outlier is a point that sits far from the rest of the data. An influential point may be an outlier, but it does not have to be. The key difference is that influence is about changing the model, while outlier is about being unusual in the data pattern.
Leverage
Leverage describes how far a point's x-value is from the center of the x-data. High leverage gives a point more power to pull the regression line. In regression, leverage is one of the main reasons a point becomes influential even if its y-value is not wildly off.
Residual
A residual is the vertical distance between an observed point and the regression line. A point can have a large residual and still not be very influential, especially if its x-value is ordinary. That is why residual size alone does not tell the whole story.
Cook's Distance
Cook's distance is a diagnostic measure that helps show how much a point affects a regression model. If removing one point would change the fitted line a lot, Cook's distance tends to flag it. It is a more direct influence check than just looking for a strange point on the graph.
A quiz item or free-response problem may show a scatterplot and ask whether one point is influential, or how the regression line changes if that point is removed. Your job is to look for leverage, residual size, and how much the slope or predictions shift. If the point has an extreme x-value and the fitted line changes a lot when it is included, that is a strong sign of influence.
You may also be asked to interpret a diagnostic like Cook's distance or compare two regression models, one with the point and one without it. Do not just label the point as an outlier and stop there. State the effect on the model, because influence is about the change in the analysis, not just the point's location.
Outliers and influential points overlap, but they are not the same thing. An outlier is far from the pattern, while an influential point is one that changes the regression result a lot. A point can be unusual without affecting the line much, and a point can be influential mainly because it has high leverage.
An influential point is a data point that changes a statistical model a lot, especially in regression.
High leverage often makes a point influential because an extreme x-value can pull the line toward itself.
A large residual does not automatically mean a point is influential, and a point can be influential even with a small residual.
In Intro to Statistics, you check influential points to decide whether a regression result is trustworthy or being overly shaped by one observation.
Diagnostic tools like Cook's distance help you spot when one point is changing the fit more than it should.
An influential point is a data point that changes a regression model noticeably when it is included or removed. In Intro to Statistics, it usually shows up in scatterplots and line of best fit problems. The point may affect the slope, intercept, correlation, or predictions.
No. An outlier is unusual because it sits far from the data pattern, but an influential point is unusual because it changes the model a lot. Some outliers are influential, and some are not. The difference matters most in regression questions.
Look for a point with high leverage or one that makes the regression line change a lot when removed. In class, you may also use Cook's distance, DFFITS, or DFBETAS. If the fitted line shifts noticeably, the point is likely influential.
Leverage measures how extreme a point's x-value is compared with the rest of the data. A point with high leverage has more ability to rotate or pull the regression line. That is why a point can strongly affect the model even if it is not far off in the y-direction.