笔记-多表查询好代码比较

多表查询一些好的代码比较，上方为案例好代码，下方为自己的代码

SELECT
    t1.university,
    t3.difficult_level,
    COUNT(t2.question_id) / COUNT(DISTINCT(t2.device_id)) as avg_answer_cnt
from
    user_profile as t1,
    question_practice_detail as t2,
    question_detail as t3
WHERE
    t1.university = '山东大学'
    and t1.device_id = t2.device_id
    and t2.question_id = t3.question_id
GROUP BY
    t3.difficult_level;
    
    上面select部分还是和我的select部分选出abc的某个字段一样，但是就在where部分
    把需要关联=的条件写在这里，使得逻辑比较清晰，不像我需要innner来inner去
    
    
 SELECT 
     a.university,
     c.difficult_level, 
     COUNT(b.device_id)/COUNT(DISTINCT b.device_id)平均答题数目 
FROM 
    user_profile a 
    INNER JOIN question_practice_detail b on a.device_id=b.device_id 
    INNER JOIN question_detail c ON b.question_id =c.question_id 
WHERE 
    a.university='山东大学' 
GROUP BY 
    c.difficult_level