DataTerminology.comNEW
Middle value that divides an ordered dataset into two equal halves.
DESCRIPTION
The median is the 50th percentile, splitting ordered observations into equal halves. It is robust to outliers and skew, often preferred for skewed distributions like incomes or latency. For even sample sizes, the median is typically the average of the two central values. Medians underpin robust measures such as the median absolute deviation. In optimization, the median minimizes absolute deviations (L1). In reporting, pairing median with interquartile range conveys central tendency and dispersion with robustness.
ALTERNATIVE DEFINITIONS
Wikipedia
The median is the value separating the higher half from the lower half of a data sample, a population, or a probability distribution.
Highlights the percentile interpretation across samples and distributions.
Khan Academy
The median is the middle number in a sorted, ascending or descending, list of numbers.
Provides a clear procedural definition for learners.
RELATED TERMS
Mode Quantile Interquartile Range Median Absolute Deviation Trimmed Mean Robust Statistics
COMPETING TERMS
KNOWLEDGE AREAS
Mathematics Data Science Data Visualization
TARGET AUDIENCE
Data Scientist Data Analyst BI Specialist Analytics Engineer
EXTERNAL SOURCES
- Wikipedia - Median
- NumPy - Median function documentation
- NIST Engineering Statistics Handbook - NIST reference on medians
- Penn State STAT Online - Median and percentiles overview
TOP QUESTIONS
When is the Median preferred over the Mean?
Use the median for skewed distributions or when outliers are present. It better reflects central location without being pulled by extremes.
Category: Requirements
How do you compute the Median for an even number of observations?
Average the two central values after sorting the data. This maintains the 50th percentile location.
Category: Misunderstanding
What is the Median Absolute Deviation (MAD)?
MAD is the median of absolute deviations from the median. It’s a robust dispersion measure often used with the median.
Category: Concept
Is Median meaningful for categorical data?
No. The median requires an ordinal or numeric order. For nominal categories, use the mode.
Category: Mistakes
How is the Median used in dashboards?
Medians summarize skewed KPIs like response time. Pair with quartiles or box plots for robust insights.
Category: Application
Can the Median be computed incrementally?
Yes, using streaming quantile algorithms (e.g., Greenwald–Khanna or t-digest) that approximate quantiles efficiently.
Category: Requirements
What is a common misunderstanding about the Median?
Some assume the median uses all values equally like the mean; instead, it depends primarily on the order, not magnitudes.
Category: Misunderstanding
How does the Median support robust modeling?
Median-based estimators and L1 objectives reduce outlier influence, improving robustness in contaminated datasets.
Category: Best practices
What visual best communicates the Median?
Box plots clearly show the median line with quartiles and potential outliers, aiding robust interpretation.
Category: Example