# Do You See False Patterns in Random Data?

• Author:
• Updated date:

## What are the random things that correlate?

The following are spurious correlations already found;

Maine divorce rate / Margarine per capita consumption

Tangled bedsheet deaths / Ski resort profits

US crude oil imports from Norway / Driver fatalities with train

Miss America age / Steam, vapors & hot object murders

iPhones / Staircase fall fatalities

New auto sales / Universal Orlando visits

## How Stats Are Used To Lie

Plotting two completely unrelated data sets together will have people inferring that the two are somehow related. That one causes a change to the other.

There is an interactive spurious correlation generator that you can discover for yourself the various illogically inferring causation from correlations.

## What is Correlation?

There is a correlation of -0.712088 in regards to 'Sunlight in Arizona' / 'Females in Pennsylvania who slipped or tripped to their death'.

The math correlation doesn't have the same meaning with what most people think it means. It does not always mean that there is a connective link or interdependence between the two data sets.

The closer the value of r is to +1, the stronger the linear relationship is, a negative number is the inverse. if r is close to zero, then there is no linear relationship.

## There Are Already 30,000 Known Spurious Correlations

Scroll to Continue

Some real life spurious correlations in the realm of mass media:

Living Next to Freeways Causes Autism (L.A. Times)

Junk food does not cause obesity

## Big Data

BlueCRUSH is the acronym of a predictive policing program that give probabilities of future crime and is short for 'Crime Reduction Utilizing Statistical History' - funny why the acronym is 'crush' - a simple coincidence or something more ominous?

The problem is that there will always be an algorithmic bias in AI because that's how it was programmed.

### China to give every citizen a 'social credit score' by 2020

"We should expect correlations that are statistically significant but ultimately meaningless to pop up all over the place once large quantities of data are available. " - Tragedy of Big Data

To see the large image of how the Chinese social credit is planned then go here where it will show the input of outstanding behavior/awards, compliance with government regulations, criminal history that will then regulate future travel privileges, career opportunities, intensity of government monitoring and access to public service projects.