/ /

$median (expression)

Definition

$median: New in version 7.0.
Returns the median (50th percentile) of an array of numeric values.

Note

Disambiguation

This page describes $median when used as an expression, which calculates the median of an array of values.

You can also use $median in these other contexts:

$median (accumulator), used in grouping to return the median value across a group of input documents.
$median (window function), used in the $setWindowFields stage to return the median value from documents in a window.

Syntax

{
   $median: {
      input: <expression>,
      method: <string>
   }
}

Command Fields

Field	Type	Necessity	Description
`input`	Array	Required	Can be either: An expression that resolves to an array. For example, `$grades`, where the document field `$grades` contains an array of numeric values. An array of expressions, each of which evaluates to a numeric value. For example, `[ "$field1", "$field2", "$field3" ]`. `$median` calculates the median across all values in the array. Non-numeric values are ignored.
`method`	String	Required	The method that MongoDB uses to calculate the median value. The method must be `'approximate'`.

Behavior

When $median is used as an expression, it always uses an accurate percentile calculation algorithm rather than an approximate one, even when method: "approximate" is specified.

This is because expressions operate over arrays of bounded size where it is acceptable to keep all of the input values in memory.

Example

The following example uses a testScores collection to demonstrate the array input form of $median.

Create the testScores collection:

db.testScores.insertMany( [
   { studentId: "2345", scores: [ 62, 81, 80, 75, 68 ] },
   { studentId: "2356", scores: [ 60, 83, 79, 72, 65 ] },
   { studentId: "2358", scores: [ 67, 82, 78, 70, 84 ] },
   { studentId: "2367", scores: [ 64, 72, 77, 58, 91 ] },
   { studentId: "2369", scores: [ 60, 53, 72, 85, 63 ] }
] )

The following pipeline uses $median to calculate the median test score for each student:

db.testScores.aggregate( [
   {
      $project: {
         _id: 0,
         studentId: 1,
         testMedians: {
            $median: {
               input: "$scores",
               method: 'approximate'
            }
         }
      }
   }
] )

[
  { studentId: '2345', testMedians: 75 },
  { studentId: '2356', testMedians: 72 },
  { studentId: '2358', testMedians: 78 },
  { studentId: '2367', testMedians: 72 },
  { studentId: '2369', testMedians: 63 }
]

In the preceding pipeline:

$project with $median finds the median value in the scores array for each document and returns it as testMedians.
Each result reflects the median score across that student's recorded test scores.

Learn More

Back

$meta

$min