Mode would probably work great for the # of kids statistic.
Just to add, mode works best for data sets with low amount of values (number of kids is usually 1-3). It completely breaks with high amount of distinct values (like $ spent).
Yes, it works best for small integer numbers, but it doesn’t provide any meaningful degree of confidence in the amount of kids, because 0,1,2,2,2,3,5 and 1,2,2,2,3,5,6 have the same mode but express very different groups.
Mode would probably work great for the # of kids statistic.
Just to add, mode works best for data sets with low amount of values (number of kids is usually 1-3). It completely breaks with high amount of distinct values (like $ spent).
Yes, it works best for small integer numbers, but it doesn’t provide any meaningful degree of confidence in the amount of kids, because 0,1,2,2,2,3,5 and 1,2,2,2,3,5,6 have the same mode but express very different groups.