Aggregation using SQL SELECT

Topic 3
Lesson 2 – SQL SELECT Aggregation Queries
Chapter 6 Murach’s MySQL

2
Aggregate functions in MySQL
AVG([ALL|DISTINCT] expression)
SUM([ALL|DISTINCT] expression)
MIN([ALL|DISTINCT] expression)
MAX([ALL|DISTINCT] expression)
COUNT([ALL|DISTINCT] expression)
COUNT(*)
GROUP_CONCAT([ALL|DISTINCT] expression)

An aggregate function can be listed in the field list of the

SELECT clause. It will compute the desired aggregation
across the tuples qualified by the query.

3
Aggregate function keywords in SQL
All aggregate functions by default exclude NULL values before
working on the data.
The DISTINCT keyword omits duplicates from the results.
The ALL keyword includes even duplicates. If nothing is
specified, the ALL is the default behavior.

id name school credits_earned credits_req

1 Smith Khoury 32 120

2 Shah D’Amore 64 128

McKim
3 Li Khoury 50 120

4
AVG and SUM in MySQL
AVG([ALL|DISTINCT] expression)
SUM([ALL|DISTINCT] expression)

• AVG and SUM can be applied to numeric expressions.

• AVG returns the average of a column (or an expression).
• SUM returns the sum of a column (or an expression)

EXAMPLE: SELECT AVG(credits_earned) FROM student;

SELECT AVG(credits_earned) AS avg_earned
FROM student;

5
MIN and MAX functions in MySQL
MIN([ALL|DISTINCT] expression)
MAX([ALL|DISTINCT] expression)

• MIN and MAX can be applied to numeric expressions.

• MIN returns the minimal value of a column (or an
expression).
• MAX returns the maximum value of a column (or an
expression)

EXAMPLE: SELECT MAX(credits_earned) AS max_earned,

MIN(credits_earned) AS min_earned
FROM student;

6
COUNT function in MySQL

COUNT([ALL|DISTINCT] expression)
COUNT(*)

COUNT counts the number of values for a field.

COUNT(*) is a special case, it counts the number of rows. It

does not eliminate rows with NULL values. It also counts
duplicate rows.
EXAMPLE: SELECT COUNT(*) FROM student;
EXERCISE: create a query to COUNT THE DISTINCT
schools in the student table.

7
GROUP_CONCAT function in MySQL

GROUP_CONCAT([ALL|DISTINCT] expression)

GROUP_CONCAT is a MYSQL specific aggregate function. It

can only be applied to text data, CHAR OR VARCHAR. It
returns a comma separated list of the values.

EXAMPLE: SELECT GROUP_CONCAT(school)

FROM students;
SELECT GROUP_CONCAT(DISTINCT school)
FROM students;

8
Examples: Aggregate functions in MySQL
SELECT COUNT(*) FROM student;
SELECT COUNT(*) AS num_students FROM student;
SELECT SUM(credits_earned) Khoury_Earned FROM
student WHERE school = ‘Khoury’;
SELECT COUNT(DISTINCT school) AS num_schools FROM
student;

id name school credits_earned credits_req

1 Smith Khoury 32 120

2 Shah D’Amore 64 128

McKim
3 Li Khoury 50 120

9
Solution: Aggregate functions in MySQL
SELECT COUNT(*) FROM students;
SELECT COUNT(*)AS num_students FROM students;
SELECT SUM(Credits_Earned) Khoury_Earned FROM
students WHERE School = ‘Khoury’;
SELECT COUNT(DISTINCT School) AS num_schools FROM
students;

id name school credits_earned credits_req

1 Smith Khoury 32 120

2 Shah D’Amore 64 128

McKim
3 Li Khoury 50 120

COUNT(*) num_students Khoury_Earned Num_schools

3 3 82 2

10
Examples 2: Aggregate functions in MySQL
SELECT MIN(credits_req) FROM students;
SELECT MAX(credits_earned)AS max_credits FROM
students;
SELECT MAX(credits_earned)AS max_credits FROM
students WHERE school = ‘Khoury’;

id name school credits_earned credits_req

1 Smith Khoury 32 120

2 Shah D’Amore 64 128

McKim
3 Li Khoury 50 120

11
Solution: Aggregate functions in MySQL
SELECT MIN(credits_req) FROM students;
SELECT MAX(credits_earned) AS max_credits FROM
students;
SELECT MAX(credits_earned)AS khoury_earned FROM
students WHERE school = ‘Khoury’;

id name school credits_earned credits_req

1 Smith Khoury 32 120

2 Shah D’Amore 64 128

McKim
3 Li Khoury 50 120

MIN(credits_req) Credits_earned Khoury_Earned

120 64 50

12
GROUP BY clause
Sometimes you want to stratify the qualifying tuples by a
specific field in the table. Use the GROUP BY clause to
define groups of tuples. It produces a result tuple for each
combination of the fields’ values in the GROUP BY list. The
GROUP BY clause has a supporting HAVING clause. The
syntax for the SELECT statement with the GROUP BY
clause is:
SELECT select_list
FROM table_source
[WHERE search_condition]
[GROUP BY group_by_list]
[HAVING search_condition]
[ORDER BY order_by_list]

13
Examples: GROUP BY clause
SELECT school FROM students GROUP BY school;
SELECT school, credits_required FROM students
GROUP BY school, credits_required;

GROUP BY can also be used to remove duplicate tuples.

id name school credits_earned credits_req

1 Smith Khoury 32 120

2 Shah D’Amore 64 128

McKim
3 Li Khoury 50 120

14
Solution: GROUP BY clause
SELECT school FROM students GROUP BY school;
SELECT school, credits_required FROM students
GROUP BY school, credits_required;

id name school credits_earned credits_req

1 Smith Khoury 32 120

2 Shah D’Amore 64 128

McKim
3 Li Khoury 50 120

School School Credits_Req

Khoury Khoury 120
D’Amore McKim D’Amore McKim 128

15
Combining Aggregation and GROUP BY

We can apply aggregations to specific groups.

Specify a query that returns the largest number of credits
earned for each school.
Student

id name school credits_earned credits_req

1 Smith Khoury 32 120

2 Shah D’Amore 64 128

McKim
3 Li Khoury 50 120

School Max(credits_earned)
Khoury 50

D’Amore McKim 64

16
Solution query: max credit by school

SELECT school, max(credits_earned) FROM student

GROUP BY school;

Student

id name school credits_earned credits_req

1 Smith Khoury 32 120

2 Shah D’Amore 64 128

McKim
3 Li Khoury 50 120

School Max(credits_earned)
Khoury 50

D’Amore McKim 64

17
HAVING clause creates a filter

The HAVING clause applies a filter to the result tuples

created by the grouping operation. It is analogous to
the WHERE clause
Student

id name school credits_earned credits_req

1 Smith Khoury 32 120

2 Shah D’Amore 64 128

McKim
3 Li Khoury 50 120

18
Example using the HAVING clause

Write a query that returns the maximum credits_earned

for each school in the students table, if the maximum
credits_earned > 60.

Student

id name school credits_earned credits_req

1 Smith Khoury 32 120

2 Shah D’Amore 64 128

McKim
3 Li Khoury 50 120

19
Solution

SELECT school, MAX(credits_earned) FROM students

GROUP BY school HAVING MAX(credits_earned) > 60;
Student

id name school credits_earned credits_req

1 Smith Khoury 32 120

2 Shah D’Amore 64 128

McKim
3 Li Khoury 50 120

GROUP BY result
School Max(credits_earned)
Khoury 50

D’Amore McKim 64

School Max(credits_earned)
Having result
D’Amore McKim 64

20
WITH ROLLUP
An extension to the GROUP BY clause, it includes summary
rows in the result for each subgroup in your groups. It
generates grand totals and subtotals for reports.
It has not been accepted into the SQL Standard but most
vendors have implemented it.

EXAMPLE: select school, sum(credits_earned) FROM

student GROUP BY school WITH ROLLUP;

EXERCISE: Apply WITH ROLLUP to the aggregate

statements written for exercise 1.

21
Solution

SELECT school, SUM(credits_earned) FROM student

GROUP BY school WITH ROLLUP;
Student

id name school credits_earned credits_req

1 Smith Khoury 32 120

2 Shah D’Amore 64 128

McKim
3 Li Khoury 50 120

GROUP BY result
School SUM(credits_earned) School SUM(credits_earned)

Khoury 82 Khoury 82

D’Amore McKim 64 D’Amore McKim 64

146
Rollup result

22
Summary
In this module you learned:
• Aggregation functions can be used in the field list
as well as in the HAVING clause
• SQL SELECT clauses: GROUP BY and HAVING
clause
• Necessary coordination between the field list and
the GROUP BY list

SQL 23
MySQL work
Let’s use the MySQL workbench to write Aggregation
statements using the ap database

Go to the website, page topic 2, and download the

sql_exercises1.sql exercise. Complete the rest of the
queries.

24
SQL work 2
Return the vendor name and vendor contact name as one
field
Return the vendor names that start with P
Count the number of vendors
Count the number of invoices per vendor
Calculate the total payment by vendor, return the vendor
name and the total
Calculate the total payment by vendor, return the vendor id
and the total payment for vendor for vendors who paid more
than $100

25