The idea behind a correlation is that you are looking for a relationship between two sets of numbers. To explain what I mean, have a look the following statements:

1. The warmer it gets, the more sales of sun protection products increase.

2. The more experience you have of using a mobile phone service, the quicker you become when using it.

The first statement is an example of what is known as a positive correlation. For example, as one variable increases, so does the other variable. Hopefully you can see that as it gets warmer the sales of sun protection products should increase too. The second statement is an example of what is known as a negative correlation whereby as one variable increases, the other variable decreases. In the second example, we would expect that as your experience of using a particular mobile phone service (such as an automated cinema listing service) increases, the rate at which you can use the service to find the particular type of information you are interested in decreases.

However, what about this statement: people who obtain A grades on psychology courses are also good at golf. This should be seen as an example of no correlation between two variables (e.g. psychology aptitude does not relate to people's golf skills).

You can calculate the strength of the relationship between two variables as long as you can pair the values. The actual correlation statistic that you calculate can range from ?1 to 0, and...