Mark Twain famously said, “There are three kinds of lies: lies, damned lies, and statistics.”
I would counter that it is much harder to lie using statistics than without statistics because statistics are so precise and easier to verify than other kinds of information. Statistics are quantifiable and it is much easier to lie using vague qualitative assertions. Most people are more influenced by anecdotes than by statistics anyhow even though the reverse should be true. Liars can come up with an anecdote to support any position and a personal story is much more persuasive for most people than a statistic. It is much harder to find and manipulate statistics that support lies. Mark Twain’s famous quote reveals more about his own ignorance than about the truthfulness of statistics. I would say that statistics make smart people smarter, and dumb people dumber.* The problem with statistics is in knowing how to interpret them and understanding their limitations. If you can do that, then statistics can only make you smarter because they reduce uncertainty about the world. Statistics add information even if it is a limited kind of information that is hard to interpret. Ignorant people interpret statistics wrong and then blame the statistics. They feel lied to.
That is why it is important to understand the statistics we use for measuring economic progress, which is principally GDP. The Great Recession that started in 2007 officially ended half way through 2009 because GDP started growing again. Most Americans felt like this was incorrect because the median income kept dropping for two more years. Americans felt like the statistics were lying because they did not understand what the GDP statistics meant. Income growth among elites was strong in the latter half of 2009 which meant that total income (GDP) was rising even though most Americans’ incomes were still falling.
This ignorance of statistics is pervasive. Even most economists seem to be unaware that they are misleading the average American by claiming that American income is growing. Most economists are either ignorant or apathetic about the wellbeing of the average American and the misuse of GDP statistics can indeed make ignorant economists act even more ignorantly than they would act without the statistic. The problem is not that the statistic lies, but that people unwittingly lie when they interpret it.
No measure is perfectly precise, and people need to be aware of its limitations when interpreting it. Statistics are less precise (and have larger margins of error) than the kinds of direct physical measurements often used in physics, but all measurements have uncertainty. Knowing the limitations of our knowledge is just as important as acquiring knowledge itself. As Will Rodgers said, “It isn’t what we don’t know that gives us trouble, it’s what we know that ain’t so.” When people are ignorant of the limitations of statistics, they make dumb conclusions.
But the bigger problem is that most people ignore statistics and just go with their feelings which are animated by stories, not statistics. For example, which is more emotional for you to read:
- My 67-year-old neighbor, Fred, died of prostate cancer yesterday while surrounded by his three kids.
- Every day, on average, 90 Americans are dying of prostate cancer this year.
The second piece of information should carry more weight for you, but it probably doesn’t and it isn’t even very engaging compared with the anecdotes (especially on video) that social marketers expert at using to manipulate your emotions which motivate your behavior. That is why anecdotes are the lies that I worry about most.
*Surprisingly, according to Google, nobody has ever written the title of this post before, but the phrase “_BLANK_ makes smart people smarter, and dumb people dumber” has been attributed to lots of other things: the internet, age, etc.
UPDATE: For an example, see: Lies (or Stupidity), Damned Lies (or Stupidity), and Statistics