如何找到拥有比所有用户平均“计数”高的“计数”数量的“成员”？

Question

I have two tables; Members and streams. I want to find the average number of streams for all members and then see which individual members (including first name) have streamed more than the average number of streams.

Currently i have this query which finds the average number of streams and the number of streams each member has;

select stream.m_ID, count(*), (select avg (count(*)) from stream group by stream.m_ID) from stream group by stream.m_ID;

I am struggling to find how to discard member who have a lower number of streams than the average number AND to join the member and stream table together

CREATE TABLE "M_5006677"."MEMBER" 
   (    "M_ID" NUMBER(2,0), 
    "FNAME" VARCHAR2(15 BYTE) NOT NULL ENABLE, 
    "MNAME" VARCHAR2(15 BYTE), 
    "LNAME" VARCHAR2(15 BYTE) NOT NULL ENABLE, 
    "EMAIL" VARCHAR2(25 BYTE) NOT NULL ENABLE, 
    "R_DATE" DATE NOT NULL ENABLE, 
    "L_DATE" DATE, 
     CONSTRAINT "PK_M_ID" PRIMARY KEY ("M_ID")

CREATE TABLE "M_5006677"."STREAM" 
   (    "STREAM_ID" NUMBER(2,0), 
    "SDATE" DATE, 
    "SPOSITION" NUMBER(5,2), 
    "F_ID" NUMBER(2,0) NOT NULL ENABLE, 
    "M_ID" NUMBER(2,0) NOT NULL ENABLE, 
    "LIVE" VARCHAR2(3 BYTE), 
     CONSTRAINT "PK_STREAM" PRIMARY KEY ("STREAM_ID")
CONSTRAINT "FK_F_ID" FOREIGN KEY ("F_ID") REFERENCE FILM
CONSTRAINT "FK_M_ID" FOREIGN KEY ("M_ID") REFERENCE MEMBER

Answer 1

I would use window functions. If you only want to count members who have ever streamed:

select m_id
from (select m_id, count(*) as cnt, avg(count(*)) over () as avg_overall
      from streams
      group by m_id
     ) s
where cnt > avg_overall;

If you want to count members who never streamed:

select m_id
from (select m_id, count(*) as cnt, avg(count(s.m_id)) over () as avg_overall
      from members m left join
           streams s
           on m.m_id = s.m_id
      group by m_id
     ) s
where cnt > avg_overall;

Answer 2

This would work for sql server:

select stream.m_ID, count(*), (select avg (count(*)) from stream group by stream.m_ID) from stream group by stream.m_ID 
having count(*) < (select avg (count(*)) from stream group by stream.m_ID)

If you want also get data about members then you should include M_ID into your query so you can join with members table like that:

select stream.m_ID, M_ID, count(*), (select avg (count(*)) from stream group by stream.m_ID) from stream group by stream.m_ID ,M_ID
having count(*) < (select avg (count(*)) from stream group by stream.m_ID)

after that you can join on M_ID with Members table:

select FNAME from Member m
join (select stream.m_ID, M_ID, count(*), (select avg (count(*)) from stream 
group by stream.m_ID) from stream group by stream.m_ID ,M_ID
    having count(*) < (select avg (count(*)) from stream group by 
stream.m_ID)) X on m.M_ID = X.M_ID