A high leverage point refers to an observation that has extreme values on one or both predictor variables, causing it to have a substantial impact on the regression line's placement within a scatterplot.
Related terms
Influential Point: An influential point is similar to a high leverage point but goes beyond its effect on just placement. It also has a significant impact on the fitted regression model's parameters, such as slope and intercept.
Cook's Distance: Cook's Distance is used to identify observations that disproportionately affect regression coefficients. It combines information about both leverage and residuals.
Multicollinearity: Multicollinearity occurs when predictor variables in a regression model are highly correlated with each other. It can amplify the effects of high leverage points, making them more influential on model estimates.