What Is the Difference between Median and Mean Salary?

Median literally means “the middle.” It refers to the middle value of a series of values arranged in numerical order, from smallest to largest. There is an equal probability of a value falling above or below it.  If there is an even number of values, then the median is calculated by averaging the two middle values. 

• For example, the median salary of the series $55,000, $162,000, $176,000, $185,000, and $193,000 is $176,000.

Half of the salaries fall below the median and half are greater than the median value. 

• The median salary of the series $55,000, $162,000, $176,000, $185,000, $193,000, and $200,000 is calculated by adding $176,000 and $185,000, then dividing by 2 to get the average of the two middle values.  In this case, the median salary is $180,500.

Again, half of the salaries fall below the median and half exceed the median value. 

The term mean refers to the average value of a dataset; specifically, the sum of the values divided by the number of values.  Using the data above, the average salary in the first example is:

• ($55,000 + $162,000 + $176,000 + $185,000 + $193,000) / 5 = $154,200

In the second example, the average salary is:

• ($55,000 + $162,000  + $176,000 + $185,000 + $193,000 + 200,000) / 6 = $161,833

When to Use Median Salary vs. Mean Salary for Compensation Benchmarking 

Depending on the nature of the data, either the mean or the median may be more useful for describing the center of the dataset.   Both estimate the location of the center of the dataset.  They may seem similar, but significant variations can occur when comparing the median salary with the mean salary.

The difference comes in when there are outliers in the data. A few people earning a much higher salary than normal can skew the results to make it look like it is normal to earn more. This is important in determining where most employees’ salaries are expected to fall, given the same or similar skillset, years of experience, education, and knowledge.  It can help determine or validate the salary midpoint and range, as well as setting salaries for new hires, offering promotions, and evaluating the salaries of current incumbents. 

The mean can be misleading if there are outliers in the data set.  In the first (very simplified) example referenced above, the median salary of the dataset provided is $176,000, while the mean salary is $154,200.  The relatively low salary of $55,000 is much lower than the other salaries in the series.  This salary may be explained by entry-level employees with minimum qualifications and little to no experience as compared to employees with many years of experience and proven performance, knowledge, skills, and abilities.  Using the median value versus the mean results in a $20,000 difference in salary.

Advantages of Using Median Salary

In general, the median is a better indicator than the mean for measuring typical values. It is not significantly changed by outliers. When a distribution is skewed, the median does a better job of describing the center of the distribution than the mean. It is best to use the median when the distribution is either skewed (i.e., there is a higher number of values at the top or bottom of the distribution) or there are outliers present.

The mean, however, is very sensitive to the most atypical values, especially very high or very low values.  If there are no outliers in the dataset, then the mean can be used since the distribution would be fairly symmetrical (a bell curve distribution).  It is best to use the mean when the distribution of the data values is symmetrical and there are no clear outliers.

For six employees with a range of salaries of $16,500, $18,000, $18,000, $20,000, $21,000, and $45,000, the mean salary is $23,083.  This salary is actually higher than five of the employees, but still much less than the highest earner.  The median salary is not as likely as the mean salary to be skewed by outliers (e.g., an extremely high or low salary that only a few people may earn).  It may not seem like much, but it can give an inaccurate view of how much most people make in the field.

When considering a larger set of numbers or industries where there are a few people who are paid extremely well, the median can give an even more inaccurate view of the typical salary. This can also occur with a few extremely low-paid outliers, which would result in a lower mean salary. Using the median approach eliminates the skewing that can happen due to the extreme salaries in the data. The median salary in the example above is $19,000, which is more in line with what the majority earns. 

This is critical information when reviewing the salaries of incumbents for job offers, especially where limited or no market data are available.  The median salary is considered more neutral than the mean salary.  A mean salary that is much higher than the median indicates salary outliers, where some employees earn much more than the overall group of employees.

Both mean and median salary calculations are useful and should be reviewed beyond the simple, straightforward mathematical equations to determine the best use of each in any given situation.  As with all compensation, it is a balance between “art” and “science” and requires a bit of detective work to investigate which is most relevant and meaningful.  Learn more about how to accurately compare mean versus median in terms of salary compensation and how mean and median salaries affect benchmarking for total compensation at SalaryExpert, powered by ERI.