Numbers Lie: How Averages Fool You

You are in a bar, it is packed, you are with 100 of your friends. The music blasts your eardrums and you see so many people you don’t know, but some you definitely want to get to know. Out of nowhere, you see Bill Gates walk in. Yes, that Bill Gates. Because of this one act, over average everyone in the bar is a millionaire, in reality, you can barely pay for your drink, but it is still true that if you averaged everyone’s wealth out, on average, everyone in the bar is a millionaire.

Information Tyranny

One of my favourite quotes is from American Cinematographer and Director Albert Maysles – “Tyranny is the deliberate removal of nuance.”

The modern world is a wash with data and information. A tsunami of ones and zeros populate and infiltrate everything we do. The algorithms are hungry little bastards who want to snap up every and all piece of intelligence they can. More is better and the gluttony is insurmountable.

This situation was seen well before now, in 1971  economist, political scientist, and computer scientist Herbert A. Simon wrote, “A wealth of information creates a poverty of attention.” We now have so much information, what do we do with it? And what does it all mean?

Make That Money

The abundance of data and the cost it takes to capture and store it leads people to believe that it has value because it exists. The amount of money being spent on a.i. is crazy. On top of that, storing the information, creating the compute power to process the questions, and further to that, the power required to run the whole operation is staggering.

Figuring out an underserved hole in that supply chain could get you very rich but I digress.

One way we can look at the data issue is averages. They are so easy to do and with massive amounts of data it is a great way to get a feel for a dataset, or in this situation a population of people.

According to the United States Census Bureau, in 2023 the average income for all full-time, year-round workers was US$61,440. For the Asian population of the United States, the average income was US$81.020.

Other reports have similar results with the Asian American median income being US$101,418 compared to the national median income of US$70,784 in 2020.

From the face of it, if you are Asian in America you are doing alright. Earning well above the average means you have less worry about where the next meal is coming from or if there will be a roof over your head.

But what about the less fortunate Asian Americans, how are they getting on?

The National Alliance to End Homelessness’ State of Homelessness: 2023 Edition shows that Asian Americans accounted for just 4.1 homeless per 10,000 people. This is by far the lowest of any racial group.

Again, it is good to be Asian in America.

Overall poverty rates in the US in 2019 were 13% for the US population but only 10% for the Asian American population according to the Pew Research Center.

Everything is coming up roses.

Those Details and The Devil Within Them

If you stopped there, you would have some view, the averages help. They are a great starting point but they don’t really tell you anything meaningful. The average is different from the distribution.

Within the Asian community, there are many subsets, even with an average poverty rate of 10% for the whole Asian community, the Burmese and the Mongolians had a poverty rate of 25%. So nearly double the average of the country, and 2.5 times that of the whole Asian Community.

The median household income of an Indian family in 2021 was $138,418 while a Burmese family is $62352, less than half of the Indian family.

If you just take the averages and don’t look at how the numbers are distributed you get a story, but it might not be accurate.

The Hard Task of Soft Skills

In a world where data is king, the real power play is understanding the information means nothing until you can understand what is meaningful within it.

We are quick to look for general trends and simple analysis. It makes sense, it tells a story and most people aren’t maths people who care about any of the deeper treasures you might find because they don’t understand the techniques you are using.

The thing is the techniques you use are irrelevant, what you are trying to do is use the data to gain insight to tell a better story. A story which uses both quantitative and qualitative approaches.

You need facts to back up your story, but you also need a story to explain the facts. Data without a story is pointless, A story without data is useless.

If you try to scratch the surface you could find some delightful insights which tell a more well-rounded and nuanced story.