My physics teacher in high school first introduced me to the Dutch equivalent of the saying “to Measure is to know.” In the current climate of anti-intellectualism a.k.a. post-truth, measuring is often attacked. Let’s have a look at measuring and its pitfalls.
To measure is to know?
First, some history. The quote “To measure is to know” is usually ascribed to Lord Kelvin, the man we named the absolute temperature scale after. As far as I could find it originated in a journal published in 1883.
Interestingly, there is an equivalent Dutch saying “meten is weten”, which is ascribed to Kamerlingh Onnes. He used it in his inauguration speech at the University of Leiden in 1882.
I don’t know who was first, if the origin lay further back in history, or if both men had the same idea independently. In any case, the gist remains the same: when you measure things you can know if they are true or not.
For example, let’s assume two people are arguing over who has the biggest smartphone. They put each phone next to a ruler and check the length and width. Tada, they know how big each of the phones is. Simple, right?
A matter of units
Looking at the above example, trouble can still arise. What if person A measures in inches and person B measures in centimeters? Their measurements will be incomparable. If one of them rounds the length and width and then calculates the surface area and the other doesn’t, the rounding errors can have a significant effect on the result.
This sounds like a trivial problem. However, it becomes less simple when applied to other terrains. For example, say you’re measuring unemployment rates in a country. What is the ‘unit’ of unemployment? People who are at home all day might be unemployed, but what about people who don’t work deliberately to take care of their children? What if they do volunteer work for eight hours a day? What if they only have a paid job for a few hours a week? You quickly walk into a quagmire of edge cases that make a simple ‘unit’ of unemployment difficult to define.
Politicians love playing with these measurement units, though. One part of post-truth politics is cherry picking facts, and by fiddling with measurement units, you can seriously mess with facts. Two politicians can use the same statistics and derive opposing results from them by twisting the definitions.
What to measure?
The second problem in measurements is what to measure. In case of the question ‘who has the bigger phone’, it seems logical to measure the surface area. However, many feel you should measure the surface area of the screen only. You could also choose to measure the volume of the phones, either by measuring width, height, and length, or by submerging them in water and measuring how much water they displace.
Again, the choices are legion. This goes for measuring other things as well. It becomes even more difficult when measuring the effectiveness of policy. Say you implement a new program to combat unemployment. You enroll people in this program, and check if they’re still unemployed a year down the line. 60% are no longer unemployed? Great, program effective! But wait: is this really true? What if we look at a control group that didn’t enroll in the program… oh dear, after a year 60% of them is also no longer unemployed. Oops.
This may, again, seem trivial, but this is what Dutch politicians did a few years back, lauding a program that would help unemployed get a job in five weeks, and then lauding the results. In fact, the results were less than if you looked at the averages of unemployed people. The only thing that made the program different were it’s high costs.
Again, this can be a dangerous tool for those wanting to distort truth.
Guesstimates
Measuring phones is all well and good, but measuring things like climate change is a lot more difficult. For example, in the Netherlands, the Methane emissions are calculated by looking at the amount of gas we pump out of the ground, how much livestock we have, etc. This data is input into a model, which then results in methane emissions.
Seems legit, right?
Wrong. Last week, a journalist wrote an article about how actual measurements of the air are showing very different results.
The problem, of course, is that emissions are a very complex matter, and it’s not possible to measure exactly how much methane or CO2 is emitted into the atmosphere in exactly what way. Part of that is always an estimate based on the data you have. A guesstimate in other words.
This does not mean measurements are useless, but it does show that it is important to always improve your measurements and models.
Trump wants to defund much of the part of NASA that measures the Earth. This will make measurements of the climate more difficult and fits better with his narrative that climate change doesn’t exist; why waste money on measurements if they’re not 100% accurate.
To measure is stupid?
Post-truth politics tell us that since measurements are often flawed, you should go with your feelings – as you can argue, with the flaws I described above.
This is actually pretty stupid. The fact that measurements are imprecise and subject to debate doesn’t mean they’re useless. To measure is to know, but to not measure is to not know.
If you don’t check your assertions, then you’re certain you won’t know their validity. You can pretend there is no truth, but there is always a real truth, even if you can’t measure it or don’t want to see it.
A phone does have a fixed size, and fudging the measurements doesn’t make it bigger or smaller. Climate change is happening, and ignoring it won’t provide people with food and water when their country becomes a desert.
In short, always be critical, but always be sure to measure as accurately as you can.
