Skip to main content
SalaryTruthBLS DATA

Median Salary

The middle point of all salaries for a given role — half of workers earn more, half earn less. More useful than average salary because it isn't skewed by extremely high or low earners.

How It Works

The median is the most commonly cited salary statistic and the best single number for understanding "typical" pay. Unlike the mean (average), the median isn't distorted by outliers. For example, if 10 software engineers earn $90,000-$130,000 and one earns $500,000, the mean salary would be inflated by the top earner, but the median would accurately reflect what a typical engineer earns. BLS reports the median as the 50th percentile (p50) in the Occupational Employment and Wage Statistics (OEWS) survey. SalaryTruth uses the BLS median as its primary salary metric because it provides the most honest answer to "what does this job actually pay?" National medians are useful starting points, but local medians can vary enormously — a software engineer's median salary in San Francisco is nearly double the national median. Always compare salaries at the metro area level and adjust for cost of living when evaluating offers across different cities.

Related Terms

  • Salary PercentileYour position in the salary distribution — the 75th percentile (p75) means you earn more than 75% of workers in the same role. BLS reports p10, p25, p50, p75, and p90.
  • Mean Salary (Average Salary)The arithmetic average of all salaries for a role — calculated by adding all salaries and dividing by the number of workers. Typically higher than the median because high earners pull the average up.
  • Wage DistributionThe full range and spread of salaries for a given occupation — from the lowest earners (p10) to the highest (p90) — revealing how pay varies by experience, location, and employer.

About This Definition

This definition is part of the SalaryTruth Salary & Career Glossary25 terms explaining compensation, salary data, and career development. All salary data from the Bureau of Labor Statistics OEWS survey.