Home » scientific method » How to create a hypothesis

How to create a hypothesis

Start here


Science starts with people looking at something in our world. Someone may notice patterns. They many wonder why things behave the way that they do. At that point they may make a hypothesis. But what exactly is a “hypothesis”?


If you’re repairing a car then a hypothesis would be your tentative explanation for why the car isn’t working. After you do a repair you then test it by seeing if the car now works.

If you’re a detective investigating a crime then your hypothesis is your tentative explanation for who did it and how they did it. After you do a careful investigation of all the evidence you can find, then you see if this hypothesis is supported by the evidence.

In science the word hypothesis has a special meaning – and it probably doesn’t mean what you think it does: “Hypothesis” doesn’t mean “educated guess.”

Vocabular: tentative – something that we accept as true for now, but we’re willing to change our minds if evidence compels us to.

Here’s a better definition

A hypothesis is a tentative explanation for a phenomenon that can be tested by experiment.

A phenomenon is any kind of observable or measurable occurrence in nature.

What are some examples of phenomenon?

* When water gets hot enough it boils

* When any metal gets hot enough, it melts, and then boils.

* When a rock is dropped, gravity pulls it down faster and faster every second

* When you pluck a guitar string, it vibrates and creates a sound

* When you hit any tightly stretched object, it vibrates and creates a sound

calvinandhobbes toast

Suppose that your car will not start. Create a hypothesis.

“My car does not start because the battery is low.”


* If the headlight switch was left on for a long time, this would result in a battery drain.

* The starter will make a certain type of sound if the battery is dead (many of you will know what I am talking about here.)

  • If the battery is dead, the voltage across the battery terminals will be much lower than normal.

Experiment: Test to see if predictions are verified or refuted

* Check whether the lights were left on.

* Insert the key and turn it; listen to the sound of the engine.

* With the right device, check the voltage of the battery.

Evaluate the results:

If predictions are verified, then our hypothesis was confirmed.

If predictions are refuted then our hypothesis must be rejected.

If that’s the case then we come up with another hypothesis, e.g.  “The starter is broken”, or “Out of gas”, etc.

How to write a hypothesis

Your hypothesis must be something you can test.

“If I do [this], then [that] will happen.”

If I [give flower seeds an organic fertilizer] then [they will sprout faster than those fed a synthetic fertilizer].


BAD – Planets move around the Sun at different speeds, because different angels push them. – This is untestable.

GOOD – Planets move around the Sun at different speeds, because at different distances from the Sun, they experience a different amount of pull from the Sun’s gravity. – This is testable.

%d bloggers like this: