Sign In for Full Access

Quick access through the institutional single sign-on Manchester Met Sign In
Skip this for now
Public Access Here

Sign In for Free Access

Login with email for free guest access to a range of Rise content
Logging You In!
Incorrect Password (Click Here to Reset)! Passwords Must Match Password must be more than 8 characters
Skip this for now
Man Met Access Here


When using inferential tests, such as Tests for Difference, hypotheses must be considered. 

A hypothesis is an assumption about how two or more variables are associated/different from one another/related. A simple hypothesis makes a prediction on how the independent variable/s will affect the dependent variable. Hypothesis statements are created prior to conducting any research and drive the future study, because one of the main aspects of inferential tests is to generalise the findings to the wider population. It makes sense to try and predict what these findings might be. Let’s say for example you are interested in crime; you want to explore whether age affects likelihood to be a victim of crime. We can develop a hypothesis from this, for example: 

Younger people are more likely to be a victim of crime than older people. 

But we must remember, our aim isn’t to just be referring to our sample, we want to know if this statement is true for the wider population. So we need to create a hypothesis that allows us to test this. 

Inferential statistical tests call for us to develop two statements, known as the Null Hypothesis and the Research Hypothesis. These statements should mirror one another, being they follow the same structure where one (the Null Hypothesis) is the negative version of the other (Research Hypothesis). For example: 

Null Hypothesis: There will not be an association between age and victimhood of crime 

Research Hypothesis: There will be an association between age and victimhood of crime

As you can see these two statements are mirroring each other as intended, simply add not to a null hypothesis to ensure appropriateness. Essentially, the null hypothesis is stating nothing will happen whilst the research hypothesis is saying ‘something’ will happen.

Two-tailed and One-tailed Hypotheses 

This is also known as a non-directional or two-tailed hypothesis. A Two-tailed hypothesis simply predicts that the independent variable will have an effect on the dependent variable, but the direction of the effect isn’t specified. In the example above, it’s stated that there will be an association between age and victimhood of crime but not what it could be. Are you predicting that younger or older people have a greater likelihood to be a victim of crime? 

Whilst a non-directional/two-tailed hypothesis doesn’t predict the exact outcome, a directional or one-tailed hypothesis does. A one-tailed directional hypothesis predicts the nature of the effect, of the independent variable, on the dependent variable. For example, younger people are more likely to be a victim of crime. 

Independent and dependent variables 

In all of the above example hypotheses they’ve included two variables, the independent and dependent variable. You must know which is which to properly construct a hypothesis. 

An independent variable (IV) can be thought of as the variable not influenced by other variables being measured, these variables are typically thought of as being the cause and they influence the dependent variable. In the example above age is the independent variable. 

A dependent variable (DV) is what you’re measuring in a statistical test and depends on other factors (such as the IV). It’s regarded as the outcome variable, showing the effect of the IV. In the above example, likelihood to be a victim of crime is the DV. 

There will be an association between age and victimhood of crime

It is sometimes possible for a variable to be an independent variable, in one hypothesis and a dependent variable in another. It’s just important that you understand what you want to measure, the IV and DVs for each hypothesis and statistical test you do. 

Let’s test your knowledge of independent and dependent variables

Observing the hypotheses below, match the hypotheses with its relevant independent and dependent variable. 


Hypotheses for Test for Difference

So far, you’ve been shown how to construct a simple hypothesis for inferential tests but when you’re doing tests for difference, your hypothesis will look slightly different. The process is the same with only one slight change of wording. When conducting a Test for Difference your hypothesis must refer to difference in replacement of association. For example, 

Instead of simply saying There will be an association between age and victimhood of crime 

You could write 

There will be an age difference in likelihood to be a victim of crime 

There will not be an age difference in likelihood to be a victim of crime. 


People’s likelihood to be a victim of crime will differ according to their age 

People’s likelihood to be a victim of crime will not differ according to their age. 

The key thing here is the word difference as you’re testing for difference. Furthermore, to reiterate the learning from this sprint, when creating hypotheses, ensure you know which variable is your IV and DV. In addition to which type of hypothesis you wish to produce, one-tailed or two-tailed. 

Now, have a go at developing your own hypotheses for a Test for Difference using the variables below. 

IV: Gender

DV: Interest in politics 

Null Hypothesis: 

Research Hypothesis: 

IV: Social Class 

DV: Quality of life  

Null Hypothesis: 

Research Hypothesis: 

IV: Religion 

DV: Attendance at place of worship

Null Hypothesis: 

Research Hypothesis:

Insert infographic about hypotheses