I have 2 datasets. One associated with Accidents and one with the vehicles involved in the same. Now 1 accident can have more than one vehicles involved. While merging these, I need a single row of accident -> vehicle and hence, I am trying to think of a strategy to do the same. For numerical variables in vehicles, I plan to take the average (like Age of driver, engine power, etc.), however, how exactly can I merge 3-4 rows of categorical variables into 1 row? Like let’s say if there are two males and two females for sex of driver, which one to choose if I am looking at frequency-based?