Lies, damned lies, and … (#1)

Okay, so I wanted to write about statistics, and I’m hopelessly unoriginal (I’m sure some people will be able to suggest a better title). This is one of those posts that started out as a rant inspired by one of my pet peeves, but hopefully it ends with some helpful reflections. We shall see!

Part I—The rant

I am heartily sick of reading ethical arguments based on statistics. For the purposes of this little grumble, I will provide a hypothetical example, but I’m sure you could come up with hundreds of real examples in the next five minutes by googling half a dozen newspapers around the world. (Aside: for those with an interest in the matter, after pondering for some time whether the verb ‘to google’ warranted a capital, I discovered this fascinating article on Wikipedia that restored my faith in the absolute absurdity of human nature. Geeks of the world unite!)

Government set to legislate on chocolate frog dilemma

The Oztian government released a new white paper this morning on chocolate frog consumption by middle-aged men.

“There’s no doubt we have a serious issue”, commented the Department of Health spokesman, Arnold Gerfrumption. “Chocolate frog consumption has been on the increase in the last decade. From being virtually unknown in the late 90s, chocofrogtion is now a common problem for many families. The figures suggest that 74 per cent of men between the ages of 35 and 55 are consuming three chocolate frogs or more a day. It’s a real problem, and the statistics tell us it’s time to act. The government will seriously consider legislation to limit the supply of chocolate frogs in the coming months”

Pat Eatemup, communications director for the action group, Freedom to Consume has lambasted the report saying it is based on faulty assumptions and faulty research methods.

Now, in any online paper in the world that allows comments, what will occur? Within hours there will be 37 (I don’t why, but it’s always 37) comments of the “I’m-a-40-year-old-male-with-a-gammy-leg-and-I-consume-frogs-responsibly-once-or-twice-a-week.-Who-made-up-these-statistcs?-Someone-should-tell-the-government-to-deal-with-real-problems” variety.

Part II—Some ‘calm’ reflection

Spoiled for targets, the humble critic searches in vain for the logical equivalent of a machine gun, but decides instead to bring in the sniper. (I.e. we’re going to look at the problems one at a time). Let’s start with our friendly armchair commentator. Statistics are statistics. They tell you about general trends in particular populations (if they have been well collected and analyzed). Your personal experience has nothing whatsoever to do with the validity or otherwise of the numbers. The 74 per cent number tells you automatically that 26 per cent of people don’t have a problem. All that the personal experience of our responsible frog-consuming friend tells anyone is that he isn’t a part of the 74 per cent. It has no bearing on whether or not the 74 per cent is an accurate figure. Can someone please teach everyone in early high school this basic fact of life? Please???

But that’s not the end of the problems because our commentator’s response was motivated, at least in part, by the faulty logic of the government position. Statements like “the statistics tell us it’s time to act” contain the heart of the problem. Statistics never tell us it’s time to act. Take this statistic: 74 per cent of middle-aged men think that elephants should be painted pink so they could be spotted more easily if they escaped from the zoo. Would any government spokesman get up and respond by saying the statistics tell us it’s time to act? The problem is again a faulty understanding of the nature of statistics. Statistics are not by their nature prescriptive, they are always descriptive. That is, they can tell you about the way things are but they do not tell you the way things ought to be. Just because 74 per cent of middle-aged men think elephants should be painted pink doesn’t mean that we should paint them pink. Similarly, the fact that 74 per cent of middle-aged men consume three or more chocolate frogs a day doesn’t tell you whether that’s a problem, or what you should do about it if it is.

The fundamental problem is, as always, sin. Where did that come from, I hear you ask. I thought the problem was faulty logic. Yes, but faulty logic is related to sin. Let me explain what I mean. As a society, we have removed God and his revelation about his creation from our thinking about everything. But this is a massive problem for a number of reasons. First, in the presence of God, I am asked to do what is right because God has made known what is right in his universe. But since the rise of scientism, we all now know that there is no God. If there is no God, then what should we do? Well, the real answer to that question is whatever we like. (The real atheist needs to accept the fruits of their thinking, but hardly ever does.) But that answer doesn’t go down too well because even the complete unbeliever still knows there is right and wrong (cf. Rom 1:32). So in the absence of any authority, people create their own authority. (After all, “because I said so” has met with very mixed success, even amongst the two-year-olds of the world!) Since, according to a scientific world, there is only one authority—‘what is’ (Empiricism)—all moral arguments now begin by quoting statistics. To summarize, because of sin (the rejection of the authority of God), we have turned to faulty logic (we pretend that ‘what is’ tells us ‘what should be’).

But here you have a second problem, which is closely related to the first. Without God, you have no vantage point from which to evaluate the statistics. A healthy doctrine of sin is vitally important when assessing statistics because only a healthy doctrine of sin tells you that the way things are is not necessarily good. What does the Bible say? My thinking, my acting, the structures of society in which I live and even the very fabric of creation, which is groaning because it has been subjected to frustration by God (Rom 8:20), are out of whack. One of the key conclusions is that any observation about ‘what is’ (and we’ve already seen that this is the only thing that statistics can give us—an appreciation of what actually is) cannot, by definition, tell you about what should be. Why? Because what is might be a terrible distortion of what should be.

For example, death is one of the few certainties in life. Does that mean death ought to be? And does the fact that it is a reality for everyone tell us anything about whether it’s acceptable to visit death on someone else? If death is the end point for everyone, what’s the problem with hastening someone else’s death? If you respond to that last question by saying, “How can you even ask such a question?” then I’m glad, but it also shows that the real issue isn’t about logic or reason. There is no logical reason that you shouldn’t ask the question. After all, death just is. The bottom line? Statistics are a false authority—a sleight of hand used by people who want us to accept their intuition about what is right and wrong.

Without God, we turn to statistics to tell us what to do. But statistics only tell us about what is. Yet in a broken, sinful world, ‘what is’ is sometimes the polar opposite of what should be. This truth has all sorts of implications, which I will come to in a future post. But for now, one conclusion is enough. Statistics are no basis for a system of government! (Or for deciding questions of morality.)