The leverage of an observation is based on how much the observation's value on the predictor variable differs from the mean of the predictor variable. The greater an observation's leverage, the more potential it has to be an influential observation.
leverage(model)
lm
leverage of the fitted model
model <- lm(mpg ~ disp + hp + wt + qsec, data = mtcars) leverage(model)#> [1] 0.14830256 0.12992351 0.06317600 0.11598460 0.17187869 0.09986745 #> [7] 0.14720085 0.12912656 0.45839085 0.13597113 0.11541159 0.10343044 #> [13] 0.04684167 0.05527722 0.20578392 0.21066328 0.19546860 0.08593044 #> [19] 0.14685385 0.14389010 0.11957800 0.11051613 0.08002229 0.10635592 #> [25] 0.19566514 0.09324142 0.15093650 0.16467565 0.23978580 0.21863752 #> [31] 0.53168134 0.07953098