Data Fairytales

The Princess & The P-Value

Anton Javelosa

29 October 2018

The data storytelling space is already saturated, which is why we decided to come up with something unique: data fairytales. Basically, if The Brothers Grimm and Alan Turing had a lovechild, this is probably what they would be reading... or writing. You never know how advanced these children can be.

A word of caution: Any resemblance to any persons living or dead is purely coincidental, or at least we try to make it appear that way.

The Princess & The P-Value

Once upon a time, in the Kingdom of Upper Management, there was a Prince Executive Officer who wanted a Data Science Princess. But she needed to be a true Data Science Princess, who understood the entire Data Value Chain.

He travelled to the Land of LinkedIn far beyond the barren JobStreet Desert and the dreaded Resumé Swamp, but there was always something wrong. Many of the Data Science Princesses he met could visualize data on Excel and Tableau, but a lot of what they showed him was useless. So, after finding no one, he returned to his Kingdom feeling very sad because he wanted a true Data Science Princess so badly.

One quarter, there was a terrible storm; it thundered and lightning-ed, and data poured down in torrents. Indeed, it was a fearful time.

In the middle of the storm somebody knocked at the office doors, and the old King of the Board himself sent to open it.

It was a Data Science Princess who stood outside, but she was in a terrible state from all the machine learning she had done the night prior. The data streamed out of her laptop and into a cloud computer; it ran on a virtual machine and three GPUs. She said that she was a true Data Science Princess and that she understood the Data Value Chain.

A Princess

This Data Science Princess cares about her p-value

‘Well we shall soon see if that is true,’ thought the old Queen Information Officer, but she said nothing. She went into the database, and used a p-value much greater than 0.05 on the raw data; then she generated twenty tables on top of the p-value, and then generated twenty charts on top of those tables. This was where the Data Science Princess was to build a predictive model that night.

In the morning they asked her how she managed.

‘Oh terribly bad!’ said the Data Science Princess. ‘I have hardly been able to build a predictive model with the data! Heaven knows what was in there. I seemed to be building a predictive model on some awful p-value that was much greater than 0.05, and the whole regression formula is screwed up—not to mention the extremely low R-squared I got in the process. It is terrible!’

They all saw at once that she must be a true Data Science Princess because she had noticed the p-value above 0.05 through twenty tables and twenty charts. Nobody but a true Data Science Princess could have had such delicate data analysis skills.

So the Prince Executive Officer invited her to run the Kingdom of Upper Management alongside him, for now he was sure that he had found a true Data Science Princess, and a new p-value much lower than 0.05 was put into the database, where it may still be seen if no one has changed it.

Now this is a true story.

Anton Javelosa
Anton is a freelance copywriter and analytics consultant at Cirrolytix. When he’s not staring at data, he can be found telling dad jokes and chasing after his two-year old daughter. He still can’t figure out how to automate potty training.


Like and share this article:

Need to contact us regarding this article?

Privacy Policy

By continuing to access this site you agree to our Privacy Policy.

More Stories