Salary Statistics

Hey guys

This is my second-to-last week here in D.C. and my fourth-to-last blog. I'll be sad to leave the D.C. area but I'll be equally happy to come back home.

Today I'm going to talk about salary in the general sense, rather than a field-by-field sense. I'll talk about the different ways salary is reported and their uses.

So when you look for a job one of the first things you look for is the salary statistics. You want to know how much you're going to make! But, they can be misleading. There are two main ways salary is reported--Average Salary and Median Salary.

Averages take the sum of all salaries in the field and divide it by the number of people employed. In datasets with low variance, averages are a good way to look at...well...an average of a bunch of different things. But, averages are sensitive to extremes. If three people in a field are making $50,000 a year, and one person is making $200,000, the average salary is $87,500. This would give job-seekers the impression that they would make, on average, that much, when that is patently untrue.

Medians work differently. Median is the exact middle value of a set of data. In a dataset with large variance, like the one previously mentioned, medians tend to be more along the lines of what one could expect to find. Because medians are dependent only on the number of points, they're far less sensitive to extremes.

For a more complete picture of the salary field, you'd use a quartile system. You'd report the salaries of the bottom 25%, the middle 50% and the top 25%. You could use median or average in this. This shows the most accurate picture of the distribution of salaries in a field and provides someone with the best estimate of what they could expect to make.