Unexpectedly Intriguing!
12 January 2010

Mr. Smith Goes to Washington - Source: National Archives The National Archives celebrated the 70th anniversary of Mr. Smith Goes to Washington with a screening of the film on Thursday, October 15, 2009. What is the distribution of salaries for the employees of the United States federal government?

To find out, we mined the data available in the Asbury Park Press' searchable database of federal employee salary records, which makes Fiscal Year 2008 salary information available for roughly 70% of all employees of the executive branch of the U.S. government, omitting data for people who work for the White House, Congress, Postal Service and various independent agencies and commissions.

But it's not user friendly. While we might have liked to be able to extract the pay data for all the employees within the database, the database only makes it possible to view 10 records at a time. Unless we were willing to pay for the entire database, which was out of the question because, well, we're cheap.

So we adopted a sampling strategy. We know that the federal government employs roughly two million civilian employees. Using a sample size calculator and setting a 95% confidence level target with a confidence interval of 1.0%, or in other words, being 95% confident that our sample is within plus-or-minus 1.0% of the actual value for a population of 2,000,000 individuals, we estimated that we would need to sample some 9,558 records at random to get a good sense of what that distribution looks like.

Confronting Math But how on Earth could we select individuals at random? The federal employee database is configured so that users can only search by Name, Agency, Job Title or Location. Of these, Agency, Job Title and Location are all things that would obviously skew the data in a non-random fashion. For example, we're not going to get a full representation if we only selected individuals who are employed in "Food Services." Or those who work for the Patent and Trademark Office. Or just those individuals who live in Washington D.C., one of the more expensive places to live in the United States, for which federal paychecks are adjusted to compensate.

We could only get a random sampling then by using the Name category, which meant that we would have to choose a name common enough for a large number of federal government employees at all agencies, employed in a large number of all job titles at all locations could be reasonably represented. So we entered "Smith."

Agent Smith (Source: Movie Villains: The Matrix) Why "Smith?" As it happens, "Smith" is the most common surname in the United States, accounting for 0.88% of all 281,421,906 Americans counted in the 2000 U.S. Census.

That returned some 8,753 records, spanning some 235 federal government agencies and locations spanning all 50 states, U.S. territories and a number of foreign nations. We've made those records available at IBM's ManyEyes site.

Best of all, those 8,753 records are enough to give us a confidence level of 95% that we're within plus or minus 1.05% of the actual distribution of income for the federal government's 2 million civilian employees!

Speaking of which, here's the distribution of income for the federal government's civilian employees, as represented by our sample of "Smiths":

Fiscal Year 2008 Percentage Distribution of Federal Employees' (Named

Running some basic numbers, we find that the median federal employee salary in Fiscal Year 2008 is $60,310, while the average is slightly higher at $68,294. The minimum listed salary for federal employees is $0, which is shared by 8 individuals named "Smith". The maximum salary is $275,000, which was earned by George B. Smith, a medical officer with the Veterans Health Service stationed in Lexington, Kentucky.

We'll be exploring more of what we find in the federal employee database in a series of posts considering the Smiths who went to work in Washington.

Image Credit: We found the image for Mr. Smith Goes to Washington at the National Archives. Meanwhile, the image of multiple "Agent Smiths" from The Matrix trilogy was extracted from Movie Villains.


About Political Calculations

Welcome to the blogosphere's toolchest! Here, unlike other blogs dedicated to analyzing current events, we create easy-to-use, simple tools to do the math related to them so you can get in on the action too! If you would like to learn more about these tools, or if you would like to contribute ideas to develop for this blog, please e-mail us at:

ironman at politicalcalculations

Thanks in advance!

Recent Posts

Indices, Futures, and Bonds

Closing values for previous trading day.

Most Popular Posts
Quick Index

Site Data

This site is primarily powered by:

This page is powered by Blogger. Isn't yours?

CSS Validation

Valid CSS!

RSS Site Feed

AddThis Feed Button


The tools on this site are built using JavaScript. If you would like to learn more, one of the best free resources on the web is available at W3Schools.com.

Other Cool Resources

Blog Roll

Market Links

Useful Election Data
Charities We Support
Shopping Guides
Recommended Reading
Recently Shopped

Seeking Alpha Certified