Stock automobiles containing dummies in the driver and front passenger seats crashed into a wall at 35 miles per hour. National Transportation Safety Board officials collected information how the crash affected the dummies. The injury variables describe the extent of head injuries, chest deceleration, and left and right femur load. The datafile also contains information on the type and safety features of each crashed car.
One way to evaluate the relationship between car features and crash injuries is using ANOVA with the various injury variables as dependent variables. Histograms of the injury variables shows that all of them are skewed right. Taking the logs of these variables results in approximately normal distributions. You may also want to take the log of the weight variable.
The variables D/P, Protection, Doors, and Size all show significant F-tests when included in an analysis of variance model with no interactions and the log of head injury criterion as the dependent variable. Year and the log of weight are not significant when included as covariates with these variables in an analysis of covariance. However, the log of weight is significant when the dependent variable is either left or right femur load.
This dataset also lends itself more advanced linear models such as MANOVA and models with various interaction terms.