我手里有一份人人网的数据，样本量大概两万。
SELECT
left(name,1) as `姓`,
count(*) as `数量` ,
concat(truncate(count(*)/(select count(*) from renren_data.student) * 100,1),'%') as `百分比`
FROM renren_data.student group by left(name,1) order by count(*) desc
limit 20| 排名 | 姓 | 数量 | 百分比 |
|---|---|---|---|
| 1 | 王 | 1955 | 8.4% |
| 2 | 李 | 1662 | 7.1% |
| 3 | 张 | 1659 | 7.1% |
| 4 | 刘 | 1219 | 5.2% |
| 5 | 陈 | 838 | 3.6% |
| 6 | 杨 | 636 | 2.7% |
| 7 | 赵 | 511 | 2.2% |
| 8 | 周 | 451 | 1.9% |
| 9 | 吴 | 418 | 1.8% |
| 10 | 孙 | 398 | 1.7% |
| 11 | 徐 | 391 | 1.6% |
| 12 | 黄 | 373 | 1.6% |
| 13 | 朱 | 294 | 1.2% |
| 14 | 马 | 275 | 1.1% |
| 15 | 郭 | 272 | 1.1% |
| 16 | 胡 | 270 | 1.1% |
| 17 | 林 | 209 | 0.9% |
| 18 | 高 | 208 | 0.9% |
| 19 | 郑 | 196 | 0.8% |
| 20 | 何 | 189 | 0.8% |
SELECT
left(uid, 1) as `uid的第一位`,
count(*) as `数量`,
concat(truncate(count(*)/(select count(*) from renren_data.student)*100 ,1),'%') as `百分比`
FROM renren_data.student
group by left(uid, 1)
order by count(*) desc
limit 10| uid的第一位 | 数量 | 百分比 |
|---|---|---|
| 2 | 16203 | 65.8% |
| 3 | 2948 | 11.9% |
| 1 | 2452 | 9.9% |
| 4 | 1270 | 5.1% |
| 5 | 781 | 3.1% |
| 7 | 733 | 2.9% |
| 6 | 674 | 2.7% |
| 8 | 512 | 2.0% |
| 9 | 450 | 1.8% |
看来不符合那个什么定理啊。