原表是个员工档案，共583人，但case when结合group by用时，写法不同，其出来的结果也不同
例1：
select distinct a.Branch,
case when kultur = ‘硕士‘ then sum(num) else ‘0‘ end as 研究生学历,
case when kultur = ‘本科‘ then sum(num) else ‘0‘ end as 大本学历,
case when kultur = ‘大专‘ then sum(num) else ‘0‘ end as 大专学历,
case when kultur = ‘中专‘ then sum(num) else ‘0‘ end as 中专学历,
case when kultur = ‘高中‘ or kultur =‘职高‘ or kultur =‘中职‘ then sum(num) else ‘0‘ end as 高中学历,
case when kultur = ‘初中‘ then sum(num) else ‘0‘ end as 初中学历,
case when kultur = ‘小学‘ then sum(num) else ‘0‘ end as 小学学历,
case when kultur = ‘‘ or kultur is null then sum(num) else ‘0‘ end as 未填写学历
from( select Branch,kultur,count(*) as num
from tb_Stuffbusic
group by branch,kultur) a
Group by a.Branch,a.kultur
结果如下图：
例2：
select Branch as 部门名称,count(branch) as 总人数,
sum(case when kultur = ‘硕士‘ then 1 else ‘0‘ end) as 研究生学历,
sum(case when kultur = ‘本科‘ then 1 else ‘0‘ end) as 大本学历,
sum(case when kultur = ‘大专‘ then 1 else ‘0‘ end) as 大专学历,
sum(case when kultur = ‘中专‘ then 1 else ‘0‘ end) as 中专学历,
sum(case when kultur = ‘高中‘ or kultur =‘职高‘ or kultur =‘中职‘ then 1 else ‘0‘ end) as 高中学历,
sum(case when kultur = ‘初中‘ then 1 else ‘0‘ end) as 初中学历,
sum(case when kultur = ‘小学‘ then 1 else ‘0‘ end) as 小学学历,
sum(case when kultur = ‘‘ or kultur is null then 1 else ‘0‘ end) as 未填写学历
from tb_Stuffbusic
group by branch
结果如下图：
例3：表中ID字段有时字段值长不够4位，但不能超过4位，故通过下面语句一次性加上。
UPDATE tb_Stuffbusic
SET ID = CASE WHEN LEN(ID) = 1 THEN ‘000‘+ID
WHEN LEN(ID) = 2 THEN ‘00‘+ID
WHEN LEN(ID) = 3 THEN ‘0‘+ID
ELSE ID
END
关于case when结合group by用时的写法举例（转）
原文：https://www.cnblogs.com/lixiaozhi/p/9198552.html
