On the impact of missing outcomes in linear regression

The linear regression model is commonly used for measuring the impact of covariates over an outcome of interest, which is typically measured through the regression coefficients of the model. However, the presence of missing outcomes can seriously affect this interpretation because we have no idea about the potential impact of the covariates on those units with missing outcomes. Here, we illustrate the consequences of the missing outcomes as the interpretation of the regression coefficients in the impact of the selection factors on the performance in the university.