MySQL计算两个子查询之间的百分比

Question

I am trying to work out the percentage of a number of students who meet certain criteria. I have 3 separate tables that I need to get data from, and then I need to get the total from one table (student) as the total of students. Then I need to use this total, to divide the COUNT of the no of students in the 2nd query.

So basically I am trying to get a count of ALL the students that are in the DB first. Then count the no of students that appear in my main query (the one returning the data). Then I need to perform the calculation that will take the noOfStudents (2) and divide by the main total (24) (no of students in DB) then *100 to give me the percentage of students who have met the criteria in the main query.

This is what I have so far:

SELECT * FROM (
(
SELECT s.firstname, s.lastname, s.RegistrationDate, s.Email, d.ReviewDate,(r.description) AS "Viva"  , COUNT(*) AS "No of Students"
FROM student s
INNER JOIN dates d 
ON s.id=d.student_identifier 
INNER JOIN reviews r
ON d.review_Identifier=r.id 
WHERE r.description = "Viva Date"
GROUP BY s.student_identifier
ORDER BY s.student_identifier)
) AS Completed
WHERE Completed.ReviewDate BETWEEN '2012-01-01' AND '2014-12-01'
;

I need to output the fields following the second SELECT and this data in turn will be displayed via PHP/HTML code on a page (the BETWEEN dates will be sent via '%s').

I wondered if I should be using 2 separate queries and then getting the value (24) from the first query to perform the calculation in the second query, but I have not been able to work out how to save as 2 separate queries and then reference the first query.

I am also not sure if it is possible to display an overall % total at the same time as outputting the individual rows that meet the criteria?

I am trying to teach myself SQL, so I apologise if I have made any glaring mistakes/assumptions in any of the above, and would appreciate any advice that's out there.

Thank you.

Answer 1

Could you do this?

SELECT COUNT(*) as TotalPopulation, 
COUNT(d.student_identifier='student') as TotalStudents, 
COUNT(d.student_identifier='student')/ count(*) *100 as Percentage of students  
from students s 
inner join dates d
on s.id = d.student_identifier
inner join reviews r
on r.id = d.review_Identifier
WHERE d.ReviewDate BETWEEN '2012-01-01' AND '2014-12-01' and r.description = 'Viva Date';

You do not need first name last name if you are just looking for counts, necessarily. This get's the count(*) of table, then whatever flag you use to identify a student in the second count(), you just had it grouped by before, which could give you wrong results considering there's much else in your select before aggregation.

You could also try:

SELECT d.student_identifier, s.firstname, s.lastname, 
s.RegistrationDate, s.Email, d.ReviewDate,(r.description) AS "Viva" 
FROM student s
INNER JOIN dates d 
ON s.id=d.student_identifier 
INNER JOIN reviews r
ON d.review_Identifier=r.id 
WHERE r.description = "Viva Date" and d.ReviewDate BETWEEN '2012-01-01' AND '2014-12-01'
ORDER BY s.student_identifier

Now, if you want to return a list, that's the second one, if you want to return a count, you would use the first query and adjust to your student_identifier.