- #[[CT230 - Database Systems I]]
- **Previous Topic:** [[SQL SELECT: Working with Strings & Subqueries]]
- **Next Topic:** [[Entity Relationship Models]]
- **Relevant Slides:** ![Topic 6 SQL_DML_aggregateFns and Group By Having.pdf](../assets/Topic_6_SQL_DML_aggregateFns_and_Group_By_Having_1664362673690_0.pdf)
-
- # Aggregate Functions #card
  card-last-interval:: -1
  card-repeats:: 1
  card-ease-factor:: 2.46
  card-next-schedule:: 2022-11-15T00:00:00.000Z
  card-last-reviewed:: 2022-11-14T16:29:15.913Z
  card-last-score:: 1
	- **Aggregate Functions** are only supported in `SELECT` clauses & `HAVING` clauses.
	- Keywords `SUM`, `AVG`, `MIN`, `MAX` work as expected and can only be applied to **numeric** data.
	- Keyword `COUNT` can be used to count the number of tuples / values / rows specified in a query.
	- We can also use mathematical operations as part of an aggregate function on **numeric** data, e.g., `+`, `-`, `*`, `/`.
- # `GROUP BY` #card
  card-last-interval:: 4
  card-repeats:: 2
  card-ease-factor:: 2.32
  card-next-schedule:: 2022-11-27T12:18:53.349Z
  card-last-reviewed:: 2022-11-23T12:18:53.350Z
  card-last-score:: 3
	- `GROUP BY <group attributes>`
	- The `GROUP BY` clause allows the grouping of rows of a table together so that all occurrences within a specified group are collected together.
	- Aggregate clauses can then be applied to groups.
	- ## Using Aggregate Functions with `GROUP BY` #card
	  card-last-interval:: 4.14
	  card-repeats:: 2
	  card-ease-factor:: 2.56
	  card-next-schedule:: 2022-11-27T15:17:10.883Z
	  card-last-reviewed:: 2022-11-23T12:17:10.883Z
	  card-last-score:: 5
		- The `GROUP BY` clause specifies the group and the aggregate function is applied to the group.
		- `COUNT(*)` can be used to *count* the number of rows (tuples) in the specified groups.
		- `SUM`, `AVG`, `MIN`, `MAX` can be used to find the sum, average, min, & max of a *numerical value* in a specified group.
	- ^^**Important:** You must `GROUP BY` **all** attributes in the `SELECT` clause *unless* they are involved in an aggregation.^^
		- This **^^wouldn't work^^** as we do not `GROUP BY` all the attributes in the `SELECT` clause - `salary` remains ungrouped.
			- ```SQL
			  SELECT dno, salary
			  FROM employee
			  GROUP BY dno -- THIS IS WRONG
			  ```
	-
- # `HAVING` #card
  card-last-interval:: -1
  card-repeats:: 1
  card-ease-factor:: 2.36
  card-next-schedule:: 2022-11-18T00:00:00.000Z
  card-last-reviewed:: 2022-11-17T20:18:23.927Z
  card-last-score:: 1
	- `HAVING <group condition>`
		- The `HAVING` clause is used in conjunction with `GROUP BY` and allows the specification of **conditions on groups**.
		- The column names in the `HAVING` clause must also appear in the `GROUP BY` list or be contained within an aggregate function, i.e., you cannot apply a `HAVING` condition to something that has not been calculated already.
		-