Grouping and Sorting Data

Objectives

• Explain the importance of grouping and sorting data stored in a database

• Group and sort data with the GROUP BY() and ORDER BY() keywords

• Craft advanced queries using aggregator functions along with sorting keywords and other conditional clauses

Grouping and Sorting Data

SQL isn't picky about how it returns data to you, based on your queries. It will simply return the relevant table rows in the order in which they exist in the table. This is often insufficient for the purposes of data analysis and organization.

How common is it to order a list of items alphabetically? Or numerically from least to greatest?

We can tell our SQL queries and aggregate functions to group and sort our data using several clauses:

• ORDER BY()

• LIMIT

• GROUP BY()

• HAVING and WHERE

• ASC/DESC

Let's take a closer look at how we use these keywords to narrow our search criteria as well as to order and group it.

Setting up the Database

Some cats are very famous, and accordingly very wealthy. Our Pets Database will have a cats table in which each cat has a name, age, breed, and net worth. Our database will also have an owners table and cats_owners join table so that a cat can have many owners and an owner can have many cats.

Creating the Database:

Create the database in your terminal with the following:

Creating the tables:

In the sqlite3> prompt in your terminal:

cats table:

CREATE TABLE cats (
id INTEGER PRIMARY KEY,
name TEXT,
age INTEGER,
breed TEXT,
net_worth INTEGER
);

owners Table:

CREATE TABLE owners (id INTEGER PRIMARY KEY, name TEXT);

cats_owners Table:

CREATE TABLE cats_owners (
cat_id INTEGER,
owner_id INTEGER
);

Inserting the values:

cats:

INSERT INTO cats (id, name, age, breed, net_worth) VALUES (1, "Maru", 3, "Scottish Fold", 1000000);
INSERT INTO cats (id, name, age, breed, net_worth) VALUES (2, "Hana", 1, "Tabby", 21800);
INSERT INTO cats (id, name, age, breed, net_worth) VALUES (3, "Grumpy Cat", 4, "Persian", 181600);
INSERT INTO cats (id, name, age, breed, net_worth) VALUES (4, "Lil\' Bub", 2, "Tortoiseshell", 2000000);

owners:

INSERT INTO owners (name) VALUES ("mugumogu");
INSERT INTO owners (name) VALUES ("Sophie");
INSERT INTO owners (name) VALUES ("Penny");

cats_owners:

INSERT INTO cats_owners (cat_id, owner_id) VALUES (2, 2);
INSERT INTO cats_owners (cat_id, owner_id) VALUES (3, 3);
INSERT INTO cats_owners (cat_id, owner_id) VALUES (1, 2);

Code Along I: ORDER BY()

Syntax

SELECT column_name, column_name
FROM table_name
ORDER BY column_name DESC, column_name ASC|DESC;

ORDER BY() will automatically sort the returned values in ascending order. Use the DESC keyword, as above, to sort in descending order.

Exercise

Imagine you're working for an important investment firm in Manhattan. The investors are interested in investing in a lucrative and popular cat. They need your help to decide which cat that will be. They want a list of famous and wealthy cats. We can do that with a basic SELECT statement:

SELECT * FROM cats WHERE net_worth > 0;

This will return:

name             age         breed          net_worth
---------------  ----------  -------------  ----------
Maru             3           Scottish Fold  1000000
Hana             1           Tabby          21800
Grumpy Cat       4           Persian        181600
Lil\' Bub         2           Tortoiseshell  2000000

Our investors are busy people though. They don't have time to manually sort through this list of cats for the best candidate. They want you to return the list to them with the cats sorted by net worth, from greatest to least.

We can do so with the following lines:

SELECT * FROM cats ORDER BY(net_worth) DESC;

This will return:

name             age         breed          net_worth
---------------  ----------  -------------  ----------
Lil\' Bub        2           Tortoiseshell  2000000
Maru             3           Scottish Fold  1000000
Grumpy Cat       4           Persian        181600
Hana             1           Tabby          21800

Code Along II: The LIMIT Keyword

Turns out our investors are very impatient. They don't want to review the list themselves, they just want you to return to them the wealthiest cat. We can accomplish this by using the LIMIT keyword with the above query:

SELECT * FROM cats ORDER BY(net_worth) DESC LIMIT 1;

Which will return:

name             age         breed          net_worth
---------------  ----------  -------------  ----------
Lil\' Bub        2           Tortoiseshell  2000000

The LIMIT keyword specifies how many of the records resulting from the query you'd like to actually return.

Code Along III: GROUP BY()

The GROUP BY() keyword is very similar to ORDER BY(). The only difference is that ORDER BY() sorts the resulting data set of basic queries while GROUP BY() sorts the result sets of aggregate functions.

Syntax

SELECT column_name, aggregate_function(column_name)
FROM table_name
WHERE column_name operator value
GROUP BY column_name;

Exercise

Let's calculate the sum of the net worth of all of the cats, grouped by owner name:

SELECT owners.name, SUM(cats.net_worth)
FROM owners
INNER JOIN cats_owners
ON owners.id = cats_owners.owner_id
JOIN cats ON cats_owners.cat_id = cats.id
GROUP BY owners.name;

This should return:

owners.name      SUM(cats.networth)
---------------  ----------
Penny            181600
Sophie           1021800

In the above query, we've implemented two joins. First, we're joining owners and cat_owners on owners.id = cats_owners.owner_id. This first joined table would look like the following if we were to query it:

owners.id  owners.name      cat_owners.cat_id  cat_owners.owner_id
---------  -----------      -----------------  -----------------
2          Sophie           2                  2
3          Penny            3                  3
2          Sophie           1                  2

With this table, we then implement a second join with cats on cats_owners.cat_id = cats.id. To better understand this, try running the provided query, but select everything rather than just the owner's name and the sum of their cats' net worth, and remove the GROUP BY line. You'll be able to see all three tables have been joined.

In our example query above, when we use the SUM(cats.net_worth) aggregator in conjunction with GROUP BY, the combination changes the way that our query behaves. Without GROUP BY, we would get a sum of the net worth of all the cats:

owners.name      SUM(cats.networth)
---------------  ----------
Sophie           1203400

By adding GROUP BY, we now get the net_worth of all cats by owner. In our original data, Sophie is the owner of Maru and Hana (100000 + 21800), while Penny is the owner of Lil' Bub (181600).

SUM looks at all of the values in the net_worth column of the cats table (or whichever column you specify in parentheses) and takes the sum of the those values, but only after those cats have been grouped.

Note: If you forget to add SUM here and just try to get cats.net_worth, you'll still group by owner, but it will only display the first cat's net worth, not the aggregate.

Code Along IV: HAVING vs WHERE clause

Suppose we have a table called employee_bonus as shown below. Note that the table has multiple entries for employees Abigail and Matthew.

employee_bonus:

Employee	Bonus
Matthew	1000
Abigail	2000
Matthew	500
Tom	700
Abigail	1250

To calculate the total bonus that each employee received, we would write a SQL statement like this:

SELECT employee, SUM(bonus) from employee_bonus group by employee;

This should return:

Employee	Bonus
Abigail	3250
Matthew	1500
Tom	700

Now, suppose we wanted to find the employees who received more than $1,000 in bonuses for the year of 2007. You might think that we could write a query like this:

BAD SQL:
SELECT employee, SUM(bonus) FROM employee_bonus
GROUP BY employee WHERE SUM(bonus) > 1000;

Unfortunately, the above will not work because the WHERE clause doesn’t work with aggregates – like SUM, AVG, MAX, etc. What we need to use is the HAVING clause. The HAVING clause was added to SQL so that we could compare aggregates to other values – just how the WHERE clause can be used with non-aggregates. Now, the correct SQL will look like this:

GOOD SQL:
SELECT employee, SUM(bonus) FROM employee_bonus
GROUP BY employee HAVING SUM(bonus) > 1000;

Difference between HAVING and WHERE clause

The difference between the HAVING and WHERE clause in SQL is that the WHERE clause can not be used with aggregates but the HAVING clause can. HAVING filters out groups of rows, created by 'GROUP BY' and WHERE filters out rows. Another way to think of it is that the HAVING clause is an additional filter to the WHERE clause.

• HAVING supports aggregate functions as it has to work with groups of rows. so for example, if there are multiple integers in a group it can filter out the groups with a low average, a high total (sum) or count how many rows are in the group.

• WHERE on the other hand deals with each row individually, so aggregate functions wouldn't work for what would you be aggregating.

Also, note syntax differences:

HAVING is after GROUP BY and WHERE is before GROUP BY changing the order will produce a syntax error.

SELECT
FROM
JOIN
  ON
WHERE
GROUP BY
HAVING
ORDER BY
LIMIT

Resources

• HAVING vs WHERE clauses

• Video Review- SQL Joins Overview

View Grouping and Sorting Data on Learn.co and start learning to code for free.