Archive for the 'background information' Category

Lance Armstrong doping case and bond defaults show challenges of probabilistic reasoning

In an earlier blog post (here) we wrote that, given the evidence available to us at the time, Lance Armstrong was probably not guilty of doping. The main line of our argument was that hundreds of doping tests from certified laboratories using accepted procedures had not found dope.  We restricted the evidence to this, in the [...]

Armstrong Affair shows USADA uninfluenced by measurement

The US Anti-Doping Agency says it has “overwhelming proof” seven-time Tour-de-France winner Lance Armstrong was cheating by doping himself with performance-enhancing substances. What is the evidence? Armstrong was tested for drugs perhaps hundreds of times during his cycling career.  Every single test pronounced him clean. That is strong evidence against doping. According to the USADA, [...]

Beliefs, prejudices and data

When you see a bunch of things moving around in a flock on a pond in the distance, making quacking sounds, and occasionally becoming airborne, you do not ask for DNA evidence to determine that they are ducks, because you already have a lot of background information about swimming ducks. If someone told you they [...]

Facebook’s new check-in measure: accuracy vs practicality

Facebook sent out a message the other day on its new way of counting “check-ins” to improve accuracy. The result of the increased “accuracy” will be a drop in the number of check-ins.  The message said: “We are revising check-in numbers on Facebook Pages to give you a more accurate picture of how people are [...]

Accident stats don’t tell you much about safety

On April 9, 1992 the Canadian Institute of Mining, Metallurgy and Petroleum awarded the Westray Mine the coveted John. T. Ryan Award. The Institute grants the award every year to the mine that had the lowest accident frequency per 200,000 hours worked during the previous calendar year in Canada. On May 9, 1992, the mine [...]

Why measurements don’t always bring opposing views together

Can measurements resolve opposing views about the truth or falsehood of controversial claims?   That indeed is one of the most important purposes of measurement.  But it does not always work out that way. Take a chemical company at which plant managers and laboratory staff around the world were at loggerheads over measurements of product brightness. [...]

More circular reasoning on statistical process control

After my last post I received a number of comments on the website KPIExperts.  Most of them completely misunderstood my point, and their misunderstanding was so fundamental that rather than reply to their comments individually I decided to write a new post.  I myself have trained thousands of people in SPC over more than two [...]

Good definitions needed for reliable measurements

The trustworthiness of any measurement depends crucially on the definition of the thing you are measuring.  This in turn depends on having adequate background information to define it. Even apparently simple things can be devilish to define: take on-time delivery. What is on-time? What counts as a delivery? Or try absenteeism. Is an employee ‘absent’ [...]

Why you cannot measure risk

In 2008, Collateralized Debt Obligations (CDO) that had been given very high ratings by the credit rating agencies collapsed. The rating agencies got blamed for faulty ratings. Once a mining association gave a coal mine the “Safest Mine Award.” One month later the mine blew up and killed 32 miners. The award is based on [...]

The numbers from your performance indicators cannot tell the whole story

Consider the figure below, which shows the KPI (key performance indicator) “inventory age in days” for 20 months. Looking at the chart raises a lot of questions. Why did inventory age shoot up by almost 40% between the 7th month and the 9th?   Why did it suddenly drop down again? Why was there much more variability in [...]