In a recent previous article, I analyzed household incomes in Alberta with some primitive statistical tools.
My inspiration was that the Alberta government was using even more primitive tools to placate its own voting base that their economic life is getting better, when it is probably not.
You’ve probably heard this common phrase before: “There are lies. There are damn lies. Then there are statistics.”
While this phrase demeans the science of statistics (which is a subset of mathematics), there is a good reason for this axiom. Far too often, practitioners of statistics use statistics to bend the data to the conclusion the practitioners had before they got the data. One big reason for using statistics incorrectly is that our society has a certain level of dishonesty. Such practitioners are manipulating data more to maintain their career than to provide a good perspective for the world to move forward. When a few too many experts use statistics in this way, it degrades the whole science of statistics.
So when a certain conclusion was reached because of statistics, it becomes easier for those with opposing opinions to attack the concept of statistics, rather than whether the statistics were properly employed or not.
In other words, western democracy will throw aside statistics when it is politically inconvenient. Those with authority will just sell their version of the world to a public that has fewer skills in statistics than I have. And I only have a “learner’s permit.”
My Learner’s Permit
When I was taking my university statistics course, I remember using a mathematical template to analyze data sets of normal distributions.
Then my professor taught us how to convert the normal distribution into a t-test and a chi-squared test. I never really understood why we used one test over the other. When writing my statistics exam, I guessed on which of these two methods is appropriate. That is probably a good reason why my mark was a minimal pass.
We were also introduced to left- and right-skewed distributions. But the math behind these distributions was much more complex than normal distributions. So any rigorous analyses of these distributions was left to a higher-level statistics course — which I did not take.
Statistics is a wonderful science. But too often, it is too easily cast aside when it doesn’t fit a political agenda.
The Consultancy
Chapter 8 of my TDG book deals with options for the TDG. One of those options is “The Consultancy,” where the TDG acquires a pool of credible experts in various fields. The main objective of The Consultancy is not to find the expert of the experts, but to collate the findings of many experts. The elected TDG representatives will not have time and energy to for this kind of investigation, so it will rely on The Consultancy.
The Consultancy should employ professional statisticians. These people should overlook the workings of the other experts — scientific, economic, humanistic fields, et al — to ensure statistics have been applied in a proper way. In essence, these experts will vet the experiments that are leading to better public policy.
Nearly all professions have some subjectivity on proper ways to conduct that profession. For example, there are two kinds of dentists. The first kind like to fix every small thing. The second kind like to wait until some things get a little bigger. Both approaches have their pros and cons — and both approaches can keep our dental health in a good place. In a like manner, we should expect statisticians to have different preferred ways to analyze the same data.
So I would expect the statisticians working for The Consultancy to know of the various accepted statistical approaches. To make their report, they could apply these different approaches. Sometimes the approaches would agree with each other. This should allow the TDG to make its decision with more certainty. Sometimes the approaches would produce different conclusions. The TDG should know about this difference, and the decision would reflect more uncertainty.
And as the TDG representatives gain some experience with the expert statisticians, the elected representatives will gain a sense of the statisticians who are better at analyzing the numbers objectively rather than promoting their biases.
It should be noted that the elected TDG bodies are not obligated to use the recommendations of The Consultancy. Rather, The Consultancy will allow the bodies to see a bigger picture with statistics to make a better decision.
Conclusion
Today’s democracies have an unreliable relationship with statisticians.
In TDG governance, statistics will be better employed to find the better solutions.
Published on Medium 2024
Winston Churchill & The Turning Point
