Quick quiz: If you randomly sampled just 5 people from a population of 20,000 or more, could you use that data to tell your stakeholders anything useful?

For most data scientists, the answer is probably no.

But according to Douglas Hubbard, author of How to Measure Anything, “you need far less data than you think.”

Here’s the proof:

If you randomly sample just five people from an organisation of any size and record the lowest and highest values of whatever you’re measuring, there’s a 93.75% chance that the true median of the entire organisation falls somewhere between those two numbers.

It’s not a trick. It’s statistics.

And it’s one of several techniques Doug has spent over 35 years applying to some of the highest-stakes decisions in business and government.

In the latest Value Boost episode of Value Driven Data Science, Doug joins me to share more of these techniques, which you can use to support high-stakes decision-making when data is scarce and every observation counts.

In just 16 minutes, you’ll learn:

1. Why a single observation reveals more than you think [01:58]
2. How Laplace’s Rule of Succession lets you estimate probabilities from tiny samples [08:25]
3. The Rule of Five and what it reveals about small sample statistics [12:08]
4. The simplest and most overlooked technique for reducing measurement uncertainty [14:07]

You don’t need more data. You just need to know what to do with what you have.

Listen now on Apple Podcasts or Spotify, or click the link below:

​Episode 110: Why You Need Less Data Than You Think​

Talk again soon,

Dr Genevieve Hayes