Welcome to our Selecting Statistical Tests guide!

This interactive guide helps you select the correct statistical test needed to analyse your data.

Ok, lets start!

What are you trying to achieve with your data.......?

(a) Do you want to predict a score or a membership of a group?

At their simplest, these techniques allow you to predict one variable's value based on the value of another variable. The variable you are trying to predict is your dependent (outcome) variable and the variable(s) you are using to predict this outcome variable are your independent (predictor) variable(s). These techniques only allow you to predict a single variable, whether it be a continuous or categorical variable; you cannot predict multiple variables.

You can have one or many predictor variables, and the type of these variables can differ (e.g. continuous, categorical, etc).

The type of questions you might be trying to answer are:

1. Can I predict 10 km run time based on an individual's aerobic capacity?
2. Can I predict exam anxiety based on knowing the number of hours spent revising?
3. Can I predict whether someone is classified as computer literate based on their performance in difference computer tasks?
4. Can I predict an individual's preferred transport (car/motorcycle) based on their responses to a risk questionnaire?

Of note, if you have some inkling that this category might be appropriate, all options here lead to statistical tests that include the word "regression" in their titles - namely, linear regression, multiple regression and logistic regression.

Is this your situation? No, keep reading on....

(b) Do you want to find out differences between groups or conditions/treatments?

There are a great many statistical tests included in this group. At their simplest, they compare the difference between two groups of related or unrelated participants. They can be far more complicated, however, incorporating multiple groups, multiple conditions/treatments and multiple dependent variables. Most often, you are comparing means between different groups.

The type of questions you might be trying to answer are:

1. What is the difference in jump height between males and females?
2. Can an exercise-training programme lead to a reduction in blood sugar levels?
3. Do stressed males and females respond differently to different stress-reduction therapies?

Is this your situation? No, keep reading on....

(c) Do you want to explore associations/relationships between variables?

These tests are conducted when you want to determine the strength of association between two or more variables. These variables can be continuous, ordinal or nominal (see our Types of Variable guide). You may be interested in answering the following types of questions:

1. Is there a relationship between height and basketball performance?
2. Is your ability to cope with stress related to your fitness level?
3. Does your level of anxiety reduce your exam ability?
4. Are males more likely to be smokers than females?

Is this your situation?