select top 10000 sc.subjectsClassname,s.subjectname,count(s.subjectname) as Num,A.teacherID,tp.jobcategory1
    from BI_SubjectLessonForSalary as A
    left join
    BI_subjects s on A.subjectID=s.subjectID
    left join
    Bi_subjectsClass sc on A.subjectsClassID=sc.subjectsClassID
    left join
    Tm_TeacherProfile tp on A.teacherID=tp.ID1
    where A.TeacherID is not null and A.teacherID>0    
    group by s.subjectname,sc.subjectsClassname,A.teacherid,tp.jobcategory1

--按什么分组就group by 什么就可以了
select top 10000 min(sc.subjectsClassname)subjectsClassname
,s.subjectname,count(s.subjectname) as Num,
min(A.teacherID)teacherID,
min(tp.jobcategory1) jobcategory1
    from BI_SubjectLessonForSalary as A
    left join
    BI_subjects s on A.subjectID=s.subjectID
    left join
    Bi_subjectsClass sc on A.subjectsClassID=sc.subjectsClassID
    left join
    Tm_TeacherProfile tp on A.teacherID=tp.ID1
    where A.TeacherID is not null and A.teacherID>0    
    group by s.subjectname

------解决方案--------------------
你group by 后面有4个字段，所以出现上述结果是没有问题的。

------解决方案--------------------
SQL code

A.teacherid,tp.jobcategory1

--你有4个字段，Groupby 就是要求你4个字段的组合都相同的，这4个字段只要有1个不相同就会是新的一组了。比如：AAAA与ABAA，这就是2组。

------解决方案--------------------
语法没有任何问题，按照楼主目前得到的结果，需要最终正确数据集是什么样的？