Wednesday, April 5, 2017

eight glasses a day

Hi,

I drink a lot of water every day. I have no idea what the optimal amount is so, like I assume many others do, I use 'eight glasses a day' as a rule of thumb and hope for the best.

On a (sort of) recent episode of More Or Less, Tim Harford explored the idea that eight glasses of water per day represents the optimal amount to drink. In short, the show explained that this benchmark was myth.

A person's optimal water intake is influenced by many things. Height, weight, environment, activity level, and many other factors contribute. With all these factors, the ideal water amount for most people tends to be more or less (!) than eight glasses.

The show dismissed a number of other commonly held beliefs. Some of these are found on this non-exhaustive list from Wikipedia. It's an interesting list for both its variety and its length.

I think there are a couple of common root causes that lead to such beliefs becoming widely held. One cause is the confirmation bias. People who believe that coffee is a diuretic will note that after each cup of coffee, they go use the bathroom. Not expecting the same effect, this same group might fail to note the bathroom trips made after an equivalently sized glass of water.

Another cause is forgetting the base rate. Napoleon is known for being short (and for being a crazy dictator). However, short is a term used relative to others in a similar group. Among French men of the time, Napoleon's height was slightly above average. Some speculate that he appeared short whenever he was accompanied by his Imperial Guard- a group selected for their superior height. But in the context of his own time, Napoleon was hardly short.

I know it is inevitable that I will sometimes believe false information. It is the age of alternative facts, no? It comes naturally to me to believe that when others speak, something relevant is being said.

But to try and fully suppress the instinct to believe what I hear is dangerous, even if I am occasionally led astray. I work best when I am immersed in a cyclical process of forming and testing hypotheses. This requires a willingness to accept new ideas.

That said, I do worry about laziness in testing the validity of what I hear. When information comes in via the news or in a book, I find it perfectly reasonable to assume someone else did the work of fact checking. With such an attitude, I am begging to be led astray.

I like listening to shows like More Or Less thanks to its constant application of the 'BS' filter. The show only looks at the facts. It's a good counter-punch against the relentless tide of suspect statistics out there. And I appreciate that when I do not have time to organize a peer-reviewed study, these fine folks over at the BBC will do some of the work for me.

The most important thing, I think, is the show's common sense approach. Their simple but sound logic suggests to listeners like me that its not so hard to fact-check. Falling victim to the confirmation bias and losing track of the base rate are not uncommon errors among even the sharpest statistical minds. But those are oversights stemming from a lapse in concentration, not a lack of intelligence. Anyone can overcome them so long as they are willing to put in the work.

Click here for a link to the ten-minute long show.