Statistics Lie

BenjaminDisraeli? said (and MarkTwain popularised) "There are three kinds of lies: lies, damned lies, and statistics." Essentially, a good presenter can take the numbers from a statistical analysis and make them say whatever they want. This applies to everything from benchmarks to productivity numbers to earnings. If they're really blowing smoke out of their ass, they'll probably be caught in their lie soon enough, but little lies pass by and a good enough con man can take your company for all you're worth.

The problem is that the devil is often in the details, and without knowing the details, interpretation of the results can be tricky. For example, I've been in situations where how things are classified makes a big difference, and those graded on statistics can start to change how things are classified. It's not that their new classifications are objectively wrong, but they are merely taking advantage of existing wiggle-room to "pad" the results. (Somewhat related: SovietShoeFactoryPrinciple)

On the origin of the quotation - http://www.york.ac.uk/depts/maths/histstat/lies.htm

"After all, facts are facts, and although we may quote one to another with a chuckle the words of the Wise Statesman, 'Lies--damned lies--and statistics,' still there are some easy figures the simplest must understand, and the astutest cannot wriggle out of." Leonard Henry Courtney, 1895

[Caveat of course that statistics do not lie, people lie about statistics. It is an example of the sad state of logic and mathematics education the world 'round.]