The study compared different methods for estimating data in panel studies. They found that some methods are more efficient than others, especially for certain types of variables. In their example about schooling, the efficiency gains were most noticeable for variables that don't change over time.