import random
import collections

PERCENTAGE = 0.2

names_all = ["albert", "bruce", "anna", "bella", "cindy", 
	"colbert", "anakin", "al gore", "carla", "cole"] * 50 # men + women, men.extend(women), etc

names_subset = random.sample(names_all, int(PERCENTAGE * len(names_all)))

names_dict = collections.defaultdict(list)
for name in names_subset:
	names_dict[name[0].lower()].append(name)

for letter in names_dict:
	names = names_dict[letter]
	names_dict[letter] = random.sample(names, int(PERCENTAGE * len(names)))

print(names_dict)