Inference: Learning from what we can't see
Much of what we know about the universe comes from inference rather than direct observation. Dark matter, exoplanets, and black holes aren’t seen directly. Scientists measure their effects and infer their existence from patterns in data. This is a core idea in data science: learning about hidden variables from observable signals.
Astronomers detect exoplanets by tracking small dips in a star’s brightness. The planet isn’t visible, but repeated patterns suggest something is orbiting. Analysts do the same when they infer customer intent from clicks or diagnose issues from log data.
Why inference matters for data science
Inference is the process of concluding evidence rather than direct proof. It matters because most real-world decisions rely on incomplete information. We rarely observe causes directly; we observe outcomes and work backward. In science and in business, inference is what allows us to move from data to understanding, even when parts of the picture remain hidden.
Inference only works if you ask the right question
Cassie Kozyrkov (Google’s first Chief Decision Scientist) warns that teams often use impressive analysis to answer the wrong question. In “Making Better Decisions with Data and AI,” she explains how you can apply perfectly valid math and still miss the real decision that matters.
For you, this is about inference. Drawing conclusions from data only helps if you're inferring about the right thing. Before trusting the output, pause and ask: What hidden assumption am I making, and what decision is this meant to inform?
How far reasoning can take you
In What If?: Serious Scientific Answers to Absurd Hypothetical Questions, Randall Munroe tackles hypothetical questions that seem impossible to answer, like “what would happen to the earth if the sun suddenly switched off” or “how could we build a LEGO bridge between London and New York?”
He approaches them using simple math, clear, reasonable assumptions, and step-by-step logical reasoning. I like the book because it shows how far careful thinking can take you, even when the question sounds absurd.