How do you write a good hypothesis?

The way to write a good hypothesis is to follow a 3 step proess.

1) Identify your variables and operationalise them.

2) Identify whether you are looking for a difference or a relationship.

3) Identify whether you are going to write a directional or non-directional hypothesis.

As long as your hypothesis includes these three things then it will be a strong statement.

Let's look at a specific example to see how we can do this:

The hypothesis we want to write is for a piece of research which is looking to see if the length of sleep impacts memory.

So let's go to step 1.

1) Our independent variable (which is the variable that we are able to change and manipulate) in this case is the ​length of sleep​, and the dependent variable (which we cannot control but is what we measure) for this piece of research is memory.​ But now we need to operationalise them. Operationalising variables means explain how we measure the variable. So for example we could operationalise length of sleep to be ​'people who slept more than 6 hours in comparison to people who slept less than 6 hours.'​ You often find that there are many ways to operationalise the dependent variable as something like memory can be measured in many ways. One way which you could operationalise the variable would be ​'number of words correctly recalled from a list.'​

So now we have both our operationalised variables, we can move on to step two.

2) We need to decide if we are looking for a difference or a relationship. A difference would be if we are directly comparing two things, whereas a relationship would be showing how one thing impacts another. If you are testing for a difference then your hypothesis will sound something like 'group A is more/less/different to group B' whereas if you are testing for a relationship you will say ​'A increases/decreases/changes as B increases.' ​​For this piece of research we are comparing people with more than 6 hours of sleep with those who had less than 6 hours of sleep so we are looking for a ​difference​. This means our hypothesis will sound like ​'people who sleep more than 6 hours will .... more/less/differently to people who slept less than 6 hours.'

Now we can move onto the final step of writing the hypothesis.

3) A hypothesis can be written as either directional (when you predict what the results will show, and so say 'A will be more than B or A will be less than B') or it can be non-directional (which is when you know that there will be a difference but do not know which one will be more or less so write 'A will be significantly different to B'). You can pick which type of hypothesis you want to write (unless the exam question specifies!) but for this example let's write a directional hypothesis. If we predict that more sleep will improve your memory we would write people who sleep more will have better memories than people who sleep less.

But now let's put everything together and write our final excellent hypothesis.

'People who sleep for more than 6 hours will recall more words correctly from a list than those who slept for less than 6 hours.