The appropriate use of multiple regression depends on being able to make four basic assumptions about the data being used to develop the regression model:

that variables are normally distributed;
that the relationship between an independent variable and the dependent variable is linear
that the variance of errors is homoscedastic; and
that there is no multicollinearity among independent variables.

Pick two of these assumptions, and describe how you would check it for a dataset that you want to use to build a regression model.