12.2 Linear Regression Inference

randomization condition

the individuals in your sample are randomly selected - part of the independence assumption

does the plot thicken? condition

the spread around the regression line should be nearly constant - part of the equal variance assumption, you can check this using a scatterplot

outlier condition

your distribution of data should not have outliers - part of the normal errors assumption

independence asumption

the errors in the underlying regression model are independent of each other

nearly normal condition

the data in your distribution should resemble a normal model - part of the normal errors assumption, can be checked using a histogram of the residuals

random residuals condition

there should be no evidence of patterns, trends, or clumping when looking at the residuals plot - part of the independence assumption

straight enough condition

the scatterplot of y against x is roughly linear - part of the linearity assumption