This past weekend was Father's Day, so I took some time to relax and read the newspaper. I found several stories that suggested interesting statistical questions. Unfortunately, the data are not available for analysis. Nevertheless, the stories are worth sharing. Over the next few days, I'll post my thoughts on several stories that have statistical content. Here's the first.

### US reveals accusations against Secret Service

This story reports that the US Secret Service has been accused multiple times since 2004 of "involvement with prostitutes, ... sexual assault, ... improper use of weapons and drunken behavior." My statistical mind wonders whether the proportion of reported accusations for the Secret Service is any different than for other security organizations, such as for US military personnel or US police officers. If the proportion of Secret Service members who have been accused of misbehavior is no different from other security personnel, why is this case generating such national interest, including an investigation in the Senate? Does the US hold the Secret Service to a higher standard than police officers and military personnel? Should it?

For that matter, I wonder how the proportion of Secret Service accusations compares to the proportion of accused misconduct of members of congress?

Share

Distinguished Researcher in Computational Statistics

Rick Wicklin, PhD, is a distinguished researcher in computational statistics at SAS and is a principal developer of SAS/IML software. His areas of expertise include computational statistics, simulation, statistical graphics, and modern methods in statistical data analysis. Rick is author of the books Statistical Programming with SAS/IML Software and Simulating Data with SAS.

1. I don't agree with the premise of this column. The argument that there is something important that can be determined about the relative proportion of cases ignores the real life significance of this event.

First, the Secret Service is charged with protecting the President. This is much more significant than the consequences of the other US services listed. There is a zero tolerance for this kind of misbehavior.
Second, the activity in question occurred in preparation for a Presidential visit to Columbia. The fact that this occurred on foreign soil is even graver.
Three, any such activity has dire consequences, regardless of its frequency or proportion.

So, making any argument based on proportion of events to judge whether the US Secret Service should be held to a higher standard is a sham use of statistics and lacks merit.

May I suggest that your blog entry be deleted? Basically, it's just an embarrassing piece.

• Thanks for you feedback and suggestion.

2. Chris Hemedinger on

Personally, I'm usually skeptical about pieces that I see in the news that cite statistics or trends, as they almost always present some sort of bias by what they omit. Sports, Finance, and Economic data have used statistics for decades and we know how to interpret those. But increasingly, political, social, and even health-related stories have included more "statistics" and I'm afraid that the numbers and stories usually don't provide enough context or data to answer all of the questions that they raise.

If every such story ended with a link to "download the data cited in this story", I'd have more confidence in what I was reading.